@peaceroad/markdown-it-figure-with-p-caption 0.17.0 → 0.19.0

package/README.md

@@ -18,7 +18,7 @@ Optionally, you can auto-number image and table caption paragraphs starting from
     2. Image `alt` text that `p7d-markdown-it-p-captions` recognizes as an image caption start (`Figure. `, `Figure 1. `, `図　`, `図1　`, etc.).
     3. Image `title` attribute that matches the same labels.
     4. Optional fallbacks (`autoAltCaption`, `autoTitleCaption`) that inject the label when the alt/title lacks one.
-        - `autoAltCaption`: `false` (default), `true`, or a string label. `true` uses locale-aware generated-label defaults from `p7d-markdown-it-p-captions`, so the label text and punctuation stay aligned with the upstream caption language data. A string uses that label verbatim. Empty alt text does not generate a fallback caption.
+        - `autoAltCaption`: `false` (default), `true`, or a string label. `true` uses locale-aware generated-label defaults from `p7d-markdown-it-p-captions`, so the label text and punctuation stay aligned with the upstream caption language data. A string is treated as a label stem that must be recognized by `p7d-markdown-it-p-captions`; setup throws if it cannot be parsed as an image caption label. This plugin appends the default joint/space unless the string already ends with punctuation such as `.` / `。` / `:`. Empty alt text does not generate a fallback caption.
         - `autoTitleCaption`: same behavior but sourced from the image `title`. It stays off by default so other plugins can keep using the `title` attribute for metadata.
 - Set `autoCaptionDetection: false` to disable the auto-caption workflow entirely.
 - Multi-image paragraphs are still wrapped as one figure when `multipleImages: true` (default). Layout-specific classes help with styling:
@@ -41,7 +41,7 @@ Optionally, you can auto-number image and table caption paragraphs starting from
 
 ### Blockquote
 
-- Captioned blockquotes (e.g., “Source. A paragraph. Ewritten immediately before or after `> ...`) become `<figure class="f-blockquote">` while keeping the original blockquote intact.
+- Captioned blockquotes (e.g., `Source. A paragraph.` written immediately before or after `> ...`) become `<figure class="f-blockquote">` while keeping the original blockquote intact.
 
 ### Video & Audio
 
@@ -50,8 +50,9 @@ Optionally, you can auto-number image and table caption paragraphs starting from
 
 ### Embedded content by iframe
 
-- Inline HTML `<iframe>` elements become `<figure class="f-video">` when they point to known video hosts (YouTube `youtube-nocookie.com`, Vimeo `player.vimeo.com`).
+- Inline HTML `<iframe>` elements become `<figure class="f-video">` when they point to known video hosts (YouTube `www.youtube.com`, `youtube.com`, `www.youtube-nocookie.com`, `youtube-nocookie.com`, Vimeo `player.vimeo.com`).
 - `<div>` wrappers are treated as iframe-type embeds only when the same HTML block contains an `<iframe ...>` tag (for example common video wrapper markup).
+- `videoWithoutCaption` classifies an iframe (including one nested in a `<div>` wrapper) as video only when its `src` host is one of the known YouTube/Vimeo hosts. A generic iframe wrapper still requires `iframeWithoutCaption`.
 - Blockquote-based social embeds (Twitter/X `twitter-tweet`, Mastodon `mastodon-embed`, Bluesky `bluesky-embed`, Instagram `instagram-media`, Tumblr `text-post-media`) are treated like iframe-type embeds when their class list contains one of those provider classes. Extra classes on the same blockquote do not block detection. By default they become `<figure class="f-img">` so the caption label behaves like an image label (Labels can also use quote labels). You can override that figure class with `figureClassThatWrapsIframeTypeBlockquote` or the global `allIframeTypeFigureClassName`.
 - `p7d-markdown-it-p-captions` ships with a `Slide.` label. When you use it (for example with Speaker Deck or SlideShare iframes), the `<figure>` wrapper automatically switches to `f-slide` (or whatever you set via `figureClassThatWrapsSlides`) so slides can get their own layout. If `allIframeTypeFigureClassName` is also configured, that class takes precedence even for slides, so you get a uniform embed wrapper without touching the slide option.
 - All other iframes fall back to `<figure class="f-iframe">` unless you override the class via `allIframeTypeFigureClassName`.
@@ -60,7 +61,8 @@ Optionally, you can auto-number image and table caption paragraphs starting from
 
 - The label inside the figcaption (the `span` element used for the label) is generated by `p7d-markdown-it-p-captions`, not by this plugin. By default the class name is formed by combining `classPrefix` with the mark name, producing names such as `f-img-label`, `f-video-label`, `f-blockquote-label`, and `f-slide-label`. 
 - With `markdown-it-attrs`, attributes attached to image-only paragraphs (for example `![...](...) {.foo #bar}`) are forwarded to the generated `<figure>`.
-- `styleProcess` controls parsing of a trailing `{...}` block from the last text token of an image-only paragraph in this plugin's own scanner. It is a narrow fallback parser, not full `markdown-it-attrs` parity, and attributes already attached to paragraph tokens by `markdown-it-attrs` are still forwarded.
+- `styleProcess` controls parsing of a trailing `{...}` block from the last text token of an image-only paragraph in this plugin's own scanner. It supports simple `.class`, `#id`, bare attributes, and quoted `key="value with spaces"` / `key='value with spaces'` pairs. It is still a narrow fallback parser, not full `markdown-it-attrs` parity, and attributes already attached to paragraph tokens by `markdown-it-attrs` are still forwarded.
+- Attribute forwarding is not sanitization. If you render untrusted Markdown, keep using an HTML sanitizer or a trusted-host policy appropriate for your application; this plugin only decides which already-parsed or narrowly parsed attributes move onto `<figure>`.
 - Attributes attached to caption paragraphs stay on the converted `<figcaption>` token after paragraph-to-figcaption conversion.
 
 ## Behavior Customization
@@ -75,7 +77,8 @@ Optionally, you can auto-number image and table caption paragraphs starting from
 
 ### Wrapping without captions
 
-- `oneImageWithoutCaption`: turn single-image paragraphs into `<figure>` elements even when no caption paragraph/auto caption is present. This is independent of automatic detection.
+- `imageOnlyParagraphWithoutCaption`: turn valid image-only paragraphs into `<figure>` elements even when no caption paragraph/auto caption is present. This includes single-image paragraphs and, when `multipleImages` is enabled, multi-image paragraphs that receive classes such as `f-img-horizontal`, `f-img-vertical`, or `f-img-multiple`. This is independent of automatic detection.
+- `oneImageWithoutCaption`: legacy alias for `imageOnlyParagraphWithoutCaption`. When both are provided, `imageOnlyParagraphWithoutCaption` wins.
 - `videoWithoutCaption`, `audioWithoutCaption`, `iframeWithoutCaption`, `iframeTypeBlockquoteWithoutCaption`: wrap the respective media blocks without caption.
 
 ### Caption text helpers (delegated to `p7d-markdown-it-p-captions`)
@@ -92,16 +95,18 @@ Every option below is forwarded verbatim to `p7d-markdown-it-p-captions`, which
 - `labelClassFollowsFigure`: mirror the resolved `<figure>` class onto the `figcaption` spans (`f-embed-label`, `f-embed-label-joint`, `f-embed-body`, etc.) when you want captions styled alongside the wrapper.
 - `figureToLabelClassMap`: extend `labelClassFollowsFigure` by mapping specific figure classes (e.g., `f-embed`) to custom caption label classes such as `caption-embed caption-social` for fine-grained control. When this map is provided and `labelClassFollowsFigure` is not set explicitly, figure-following mode is enabled automatically.
 - `labelPrefixMarker`: allow a leading marker before labels (string or array, e.g., `*Figure. ...`). Arrays are limited to two markers; extras are ignored.
+- `languages`: optional available caption-recognition catalogs delegated to `p7d-markdown-it-p-captions` (default: `['en', 'ja']`). Most users can leave this unset. Set it only when you want to restrict or extend which labels can be recognized (for example English `Figure.` and Japanese `図　`) and which catalogs are available for generated fallback labels. It is separate from the active locale used to choose among those available catalogs.
 - Automatic image-label fallback text and punctuation (`Figure. `, `図　`, etc.) are generated from `p7d-markdown-it-p-captions` locale metadata, not from a local hardcoded map in this plugin.
-- `preferredLanguages`: optional tie-break order for generated fallback labels. When omitted, this plugin derives the order once per render from `env.preferredLanguages`, `env.lang` / `env.locale`, then a cheap document-script heuristic that skips a leading hyphen-fenced frontmatter block (`---` or longer, spaces allowed before newline), and finally the raw `languages` order.
+- Generated fallback label tie-break is resolved once per render. Prefer passing the active locale through `env.locale` or `env.preferredLocales`. Compatibility fallbacks are `preferredLanguages`, `env.preferredLanguages`, `env.lang`, and `env.language`. If none of those selects an available catalog, this plugin finally uses a cheap document-script heuristic that skips a leading hyphen-fenced frontmatter block (`---` or longer, spaces allowed before newline), then falls back to the raw `languages` order. This tie-break only affects generated fallback labels; it does not change the caption-recognition dictionaries selected by `languages`. Compatibility note: for generated fallback labels, `env.locale` / `env.preferredLocales` intentionally take precedence over the legacy `preferredLanguages` option so a shared `md` instance can render different documents with different active locales.
 
 ### Automatic numbering
 
 - `autoLabelNumberSets`: enable numbering per media type. Pass an array such as `['img']`, `['table']`, or `['img', 'table']`.
 - `autoLabelNumber`: shorthand for turning numbering on for both images and tables without passing the array yourself. Provide `autoLabelNumberSets` explicitly (e.g., `['img']`) when you need finer control—the explicit array always wins.
+- Do not pass p-captions' lower-level `setFigureNumber` option to this plugin. Figure numbering is owned by `autoLabelNumber` / `autoLabelNumberSets`; `setFigureNumber` is rejected during setup to prevent two numbering systems from mutating the same caption.
 - Counters start at `1` near the top of the document and increment sequentially per media type. Figures and tables keep independent counters even when mixed together.
-- The counter only advances when a real caption exists (paragraph, auto-detected alt/title, or fallback text). Figures emitted solely because of `oneImageWithoutCaption` stay unnumbered.
-- Manual numbers inside the caption text (e.g., `Figure 5.`) always win. The plugin updates its internal counter so the next automatic number becomes `6`. This applies to captions sourced from paragraphs, auto detection, and fallback captions.
+- The counter only advances when a real caption exists (paragraph, auto-detected alt/title, or fallback text). Figures emitted solely because of `imageOnlyParagraphWithoutCaption` / `oneImageWithoutCaption` stay unnumbered.
+- Manual numbers inside the caption text (e.g., `Figure 5.`) always win. The plugin reads the exact `captionDecision.number` supplied by `p7d-markdown-it-p-captions` and updates its decimal counter so the next automatic number becomes `6`. Explicit compound/alphanumeric numbers (e.g., `Figure A.` or `Figure A.5.`) are preserved without appending an automatic number and do not seed the decimal counter. This applies to captions sourced from paragraphs, auto detection, and fallback captions.
 
 ## Basic Usage
 
@@ -128,12 +133,12 @@ Auto label numbering for images and tables.
 ```js
 const figureOption = {
   // Opinionated defaults
-  oneImageWithoutCaption: true,
+  imageOnlyParagraphWithoutCaption: true,
   videoWithoutCaption: true,
   audioWithoutCaption: true,
   iframeWithoutCaption: true,
   iframeTypeBlockquoteWithoutCaption: true,
-  removeUnnumberedLabelExceptMarks: ['blockquote'], // keep “Quote. Elabels even when unnumbered
+  removeUnnumberedLabelExceptMarks: ['blockquote'], // keep `Quote.` labels even when unnumbered
   allIframeTypeFigureClassName: 'f-embed', // apply a uniform class to every iframe-style embed
   autoLabelNumber: true,
 
@@ -147,7 +152,7 @@ If there is no label number, the label will also be deleted.
 
 ```js
 const figureOption = {
-  oneImageWithoutCaption: true,
+  imageOnlyParagraphWithoutCaption: true,
   videoWithoutCaption: true,
   audioWithoutCaption: true,
   iframeWithoutCaption: true,
@@ -175,7 +180,7 @@ const md = mdit({ html: true }).use(mditFigureWithPCaption, figureOption)
 [HTML]
 <p><img src="figure.jpg" alt="A single cat"></p>
 
-<!-- Above: If oneImageWithoutCaption is true, this img element has wrapped into figure element without caption. -->
+<!-- Above: If imageOnlyParagraphWithoutCaption (or its legacy alias oneImageWithoutCaption) is true, this img element has wrapped into figure element without caption. -->
 
 
 [Markdown]
@@ -359,6 +364,8 @@ Slide. A Speaker Deck.
 
 ### Auto alt/title detection
 
+The consumed `alt` or `title` value is cleared only after the image paragraph is confirmed to be wrappable. Tight-list images and image paragraphs with trailing non-image text keep their original accessibility attributes and baseline rendering.
+
 ```
 [Markdown]
 ![Figure. A cat.](cat.jpg)
@@ -440,7 +447,7 @@ A paragraph.
 
 ### Styles
 
-This example uses `classPrefix: 'custom'` and leaves `styleProcess: true` so a trailing `{.notice}` block moves onto the `<figure>` wrapper. This fallback only handles the final trailing attrs block on an image-only paragraph; for broader attrs syntax support, keep using `markdown-it-attrs`.
+This example uses `classPrefix: 'custom'` and leaves `styleProcess: true` so a trailing `{.notice}` block moves onto the `<figure>` wrapper. This fallback only handles the final trailing attrs block on an image-only paragraph; it supports quoted values with spaces, but for broader attrs syntax support, keep using `markdown-it-attrs`.
 
 ```
 [Markdown]
@@ -456,7 +463,7 @@ Figure. Highlighted cat.
 
 ### Automatic detection fallbacks
 
-`autoCaptionDetection` combined with `autoAltCaption` / `autoTitleCaption` can still generate caption text even when the original alt/title lacks labels, as long as the alt/title body is non-empty. The corresponding attributes are cleared after conversion so the figcaption becomes the canonical source. When these fallbacks are `true`, the generated label text and punctuation come from `p7d-markdown-it-p-captions` locale metadata rather than a local hardcoded map.
+`autoCaptionDetection` combined with `autoAltCaption` / `autoTitleCaption` can still generate caption text even when the original alt/title lacks labels, as long as the alt/title body is non-empty. The corresponding attributes are cleared after conversion so the figcaption becomes the canonical source. When these fallbacks are `true`, the generated label text and punctuation come from `p7d-markdown-it-p-captions` locale metadata rather than a local hardcoded map. When these fallbacks are strings, the string must be a label stem recognized as an image caption label by `p7d-markdown-it-p-captions`; invalid strings fail during plugin setup instead of producing a stray paragraph.
 
 ```
 [Markdown]
@@ -498,7 +505,7 @@ $ pwd
 
 ### Captionless conversion toggles
 
-If `oneImageWithoutCaption` is enabled, a single image paragraph will be wrapped with `<figure class="f-img">` even without a caption.
+If `imageOnlyParagraphWithoutCaption` (or the legacy alias `oneImageWithoutCaption`) is enabled, a single image paragraph will be wrapped with `<figure class="f-img">` even without a caption. Multi-image image-only paragraphs can also be wrapped, in which case the normal layout classes such as `f-img-horizontal`, `f-img-vertical`, or `f-img-multiple` are used.
 
 ```
 [Markdown]
@@ -510,7 +517,7 @@ If `oneImageWithoutCaption` is enabled, a single image paragraph will be wrapped
 </figure>
 ```
 
-If `videoWithoutCaption` is enabled, an `iframe` pointing to a known video host (such as YouTube or video elements) will be wrapped with `<figure class="f-video">`.
+If `videoWithoutCaption` is enabled, `<video>` elements and iframes pointing to known video hosts (such as `www.youtube.com`, `youtube.com`, `www.youtube-nocookie.com`, or Vimeo) will be wrapped with `<figure class="f-video">`.
 
 ```
 [Markdown]

package/embeds/detect.js

@@ -6,26 +6,38 @@ import {
 
 const htmlRegCache = new Map()
 const openingClassAttrReg = /^<[^>]*?\bclass=(?:"([^"]*)"|'([^']*)')/i
-const openingSrcAttrReg = /^<[^>]*?\bsrc=(?:"([^"]*)"|'([^']*)')/i
-const endBlockquoteScriptReg = /<\/blockquote> *<script[^>]*?><\/script>$/
+const iframeSrcAttrReg = /<iframe\b[^>]*?\bsrc=(?:"([^"]*)"|'([^']*)')/i
+const endBlockquoteScriptReg = /<\/blockquote> *<script[^>]*?><\/script>$/i
+const targetHtmlHintReg = /<(?:video|audio|iframe|blockquote|div)\b/i
+const blueskyEmbedHintReg = /bluesky-embed/i
+const videoTagHintReg = /<video\b/i
+const audioTagHintReg = /<audio\b/i
+const iframeTagHintReg = /<iframe\b/i
+const blockquoteTagHintReg = /<blockquote\b/i
+const divTagHintReg = /<div\b/i
 const iframeTagReg = /<iframe(?=[\s>])/i
 
 const getHtmlReg = (tag) => {
   const cached = htmlRegCache.get(tag)
   if (cached) return cached
   const regexStr = `^<${tag} ?[^>]*?>[\\s\\S]*?<\\/${tag}>(\\n| *?)(<script [^>]*?>(?:<\\/script>)?)? *(\\n|$)`
-  const reg = new RegExp(regexStr)
+  const reg = new RegExp(regexStr, 'i')
   htmlRegCache.set(tag, reg)
   return reg
 }
 
 const getHtmlDetectionHints = (content) => {
-  const hasBlueskyHint = content.indexOf('bluesky-embed') !== -1
-  const hasVideoHint = content.indexOf('<video') !== -1
-  const hasAudioHint = content.indexOf('<audio') !== -1
-  const hasIframeHint = content.indexOf('<iframe') !== -1
-  const hasBlockquoteHint = content.indexOf('<blockquote') !== -1
-  const hasDivHint = content.indexOf('<div') !== -1
+  const source = typeof content === 'string' ? content : ''
+  const hasTargetHtmlHint = targetHtmlHintReg.test(source)
+  const hasBlueskyHint = blueskyEmbedHintReg.test(source)
+  if (!hasTargetHtmlHint && !hasBlueskyHint) {
+    return null
+  }
+  const hasVideoHint = videoTagHintReg.test(source)
+  const hasAudioHint = audioTagHintReg.test(source)
+  const hasIframeHint = iframeTagHintReg.test(source)
+  const hasBlockquoteHint = blockquoteTagHintReg.test(source)
+  const hasDivHint = divTagHintReg.test(source)
   return {
     hasBlueskyHint,
     hasVideoHint,
@@ -33,21 +45,10 @@ const getHtmlDetectionHints = (content) => {
     hasIframeHint,
     hasBlockquoteHint,
     hasDivHint,
-    hasIframeTag: hasIframeHint || (hasDivHint && iframeTagReg.test(content)),
+    hasIframeTag: hasIframeHint || (hasDivHint && iframeTagReg.test(source)),
   }
 }
 
-const hasAnyHtmlDetectionHint = (hints) => {
-  return !!(
-    hints.hasBlueskyHint ||
-    hints.hasVideoHint ||
-    hints.hasAudioHint ||
-    hints.hasIframeHint ||
-    hints.hasBlockquoteHint ||
-    hints.hasDivHint
-  )
-}
-
 const appendHtmlBlockNewlineIfNeeded = (token, hasTag) => {
   if ((hasTag[2] && hasTag[3] !== '\n') || (hasTag[1] !== '\n' && hasTag[2] === undefined)) {
     token.content += '\n'
@@ -89,6 +90,13 @@ const getOpeningAttrValue = (content, reg) => {
   return match[1] || match[2] || ''
 }
 
+const getAttrValue = (content, reg) => {
+  if (typeof content !== 'string') return ''
+  const match = content.match(reg)
+  if (!match) return ''
+  return match[1] || match[2] || ''
+}
+
 const hasKnownBlockquoteEmbedClass = (content) => {
   const classAttr = getOpeningAttrValue(content, openingClassAttrReg)
   if (!classAttr) return false
@@ -105,14 +113,14 @@ const hasKnownBlockquoteEmbedClass = (content) => {
 }
 
 const isKnownVideoIframe = (content) => {
-  const src = getOpeningAttrValue(content, openingSrcAttrReg)
+  const src = getAttrValue(content, iframeSrcAttrReg)
   if (!src || src.slice(0, 8).toLowerCase() !== 'https://') return false
   const slashIndex = src.indexOf('/', 8)
   const host = (slashIndex === -1 ? src.slice(8) : src.slice(8, slashIndex)).toLowerCase()
   return VIDEO_IFRAME_HOSTS.has(host)
 }
 
-const detectHtmlTagCandidate = (tokens, token, startIndex, detector, hints, result) => {
+const detectHtmlTagCandidate = (tokens, token, startIndex, detector, hints) => {
   if (detector.requiresIframeTag && !hints.hasIframeTag) return ''
   const hasTagHint = !!(detector.hintKey && hints[detector.hintKey])
   const allowBlueskyFallback = detector.candidate === 'blockquote' && hints.hasBlueskyHint
@@ -122,9 +130,6 @@ const detectHtmlTagCandidate = (tokens, token, startIndex, detector, hints, resu
   if (!hasTag && !isBlueskyFallback) return ''
   if (hasTag) {
     appendHtmlBlockNewlineIfNeeded(token, hasTag)
-    if (detector.treatAsVideoIframe) {
-      result.isVideoIframe = true
-    }
     return detector.matchedTag || detector.candidate
   }
   consumeBlockquoteEmbedScript(tokens, token, startIndex)
@@ -136,13 +141,16 @@ const resolveHtmlWrapWithoutCaption = (matchedTag, result, htmlWrapWithoutCaptio
   if (matchedTag === 'blockquote') {
     return !!(result.isIframeTypeBlockquote && htmlWrapWithoutCaption.iframeTypeBlockquote)
   }
+  if (matchedTag === 'iframe' && result.isVideoIframe) {
+    return !!(htmlWrapWithoutCaption.video || htmlWrapWithoutCaption.iframe)
+  }
   return !!htmlWrapWithoutCaption[matchedTag]
 }
 
 export const detectHtmlFigureCandidate = (tokens, token, startIndex, htmlWrapWithoutCaption) => {
   if (!token || token.type !== 'html_block') return null
   const hints = getHtmlDetectionHints(token.content)
-  if (!hasAnyHtmlDetectionHint(hints)) return null
+  if (!hints) return null
 
   const result = {
     isVideoIframe: false,
@@ -151,7 +159,7 @@ export const detectHtmlFigureCandidate = (tokens, token, startIndex, htmlWrapWit
 
   let matchedTag = ''
   for (let i = 0; i < HTML_EMBED_CANDIDATES.length; i++) {
-    matchedTag = detectHtmlTagCandidate(tokens, token, startIndex, HTML_EMBED_CANDIDATES[i], hints, result)
+    matchedTag = detectHtmlTagCandidate(tokens, token, startIndex, HTML_EMBED_CANDIDATES[i], hints)
     if (matchedTag) break
   }
   if (!matchedTag) return null

package/embeds/providers.js

@@ -9,7 +9,6 @@ export const HTML_EMBED_CANDIDATES = Object.freeze([
     hintKey: 'hasDivHint',
     requiresIframeTag: true,
     matchedTag: 'iframe',
-    treatAsVideoIframe: true,
   },
 ])
 
@@ -22,6 +21,9 @@ export const BLOCKQUOTE_EMBED_CLASS_NAMES = new Set([
 ])
 
 export const VIDEO_IFRAME_HOSTS = new Set([
+  'www.youtube.com',
+  'youtube.com',
   'www.youtube-nocookie.com',
+  'youtube-nocookie.com',
   'player.vimeo.com',
 ])

package/index.js

@@ -13,14 +13,18 @@ import { detectHtmlFigureCandidate } from './embeds/detect.js'
 const imageAttrsReg = /^ *\{(.*?)\} *$/
 const classAttrReg = /^\./
 const idAttrReg = /^#/
-const attrParseReg = /^(.*?)="?(.*)"?$/
 const sampLangReg = /^ *(?:samp|shell|console)(?:(?= )|$)/
 const asciiLabelReg = /^[A-Za-z]/
+const attrNameReg = /^[^\s=]+$/
+const labelTrailingJointReg = /[.\u3002\uff0e:：]\s*$/
 const CHECK_TYPE_TOKEN_MAP = {
   table_open: 'table',
   pre_open: 'pre',
   blockquote_open: 'blockquote',
 }
+const hasOwnOption = (option, name) => !!(
+  option && Object.prototype.hasOwnProperty.call(option, name)
+)
 const escapeRegExp = (value) => value.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')
 const normalizeLanguageCode = (value) => {
   if (value === null || value === undefined) return ''
@@ -29,28 +33,47 @@ const normalizeLanguageCode = (value) => {
   const separatorIndex = normalized.search(/[-_]/)
   return separatorIndex === -1 ? normalized : normalized.slice(0, separatorIndex)
 }
+const appendAvailableLanguage = (target, lang, availableLanguages) => {
+  if (!lang) return false
+  if (availableLanguages.indexOf(lang) === -1) return false
+  if (target.indexOf(lang) !== -1) return false
+  target.push(lang)
+  return true
+}
 const normalizePreferredLanguages = (value, availableLanguages) => {
   if (!Array.isArray(availableLanguages) || availableLanguages.length === 0) return []
-  const source = typeof value === 'string' ? [value] : (Array.isArray(value) ? value : [])
-  if (source.length === 0) return []
-  const allowed = new Set(availableLanguages)
   const languages = []
-  const seen = new Set()
+  if (typeof value === 'string') {
+    appendAvailableLanguage(languages, normalizeLanguageCode(value), availableLanguages)
+    return languages
+  }
+  const source = Array.isArray(value) ? value : []
+  if (source.length === 0) return languages
   for (let i = 0; i < source.length; i++) {
     const lang = normalizeLanguageCode(source[i])
-    if (!lang || seen.has(lang) || !allowed.has(lang)) continue
-    seen.add(lang)
-    languages.push(lang)
+    appendAvailableLanguage(languages, lang, availableLanguages)
   }
   return languages
 }
-const prioritizeLanguage = (languages, preferredLanguage) => {
-  if (!preferredLanguage || languages.length === 0) return languages.slice()
+const prioritizeLanguages = (languages, preferredLanguages) => {
+  if (!Array.isArray(languages) || languages.length === 0) return []
+  if (typeof preferredLanguages === 'string') {
+    if (!preferredLanguages || languages.indexOf(preferredLanguages) === -1) return languages
+    if (languages[0] === preferredLanguages) return languages
+    const prioritized = [preferredLanguages]
+    for (let i = 0; i < languages.length; i++) {
+      appendAvailableLanguage(prioritized, languages[i], languages)
+    }
+    return prioritized
+  }
+  if (!Array.isArray(preferredLanguages) || preferredLanguages.length === 0) return languages
   const prioritized = []
-  prioritized.push(preferredLanguage)
+  for (let i = 0; i < preferredLanguages.length; i++) {
+    appendAvailableLanguage(prioritized, preferredLanguages[i], languages)
+  }
+  if (prioritized.length === 0) return languages
   for (let i = 0; i < languages.length; i++) {
-    if (languages[i] === preferredLanguage) continue
-    prioritized.push(languages[i])
+    appendAvailableLanguage(prioritized, languages[i], languages)
   }
   return prioritized
 }
@@ -129,34 +152,49 @@ const resolvePreferredLanguagesForState = (state, opt) => {
   ) ? opt.markRegState.languages : []
   if (availableLanguages.length === 0) return []
 
-  const explicitPreferred = opt && Array.isArray(opt.preferredLanguages)
-    ? opt.preferredLanguages
-    : []
-  if (explicitPreferred.length > 0) return explicitPreferred
-
   const optionLanguages = opt && Array.isArray(opt.normalizedOptionLanguages)
     ? opt.normalizedOptionLanguages
     : []
   const baseLanguages = optionLanguages.length > 0 ? optionLanguages : availableLanguages
   const env = state && state.env ? state.env : null
-  const envPreferred = normalizePreferredLanguages(env && env.preferredLanguages, availableLanguages)
-  if (envPreferred.length > 0) return envPreferred
 
-  const envLanguage = normalizeLanguageCode(env && (env.preferredLanguage || env.lang || env.language || env.locale))
+  const envLocale = normalizeLanguageCode(env && env.locale)
+  if (envLocale && baseLanguages.indexOf(envLocale) !== -1) {
+    return prioritizeLanguages(baseLanguages, envLocale)
+  }
+
+  const envPreferredLocales = normalizePreferredLanguages(env && env.preferredLocales, baseLanguages)
+  if (envPreferredLocales.length > 0) {
+    return prioritizeLanguages(baseLanguages, envPreferredLocales)
+  }
+
+  const explicitPreferred = opt && Array.isArray(opt.preferredLanguages)
+    ? opt.preferredLanguages
+    : []
+  if (explicitPreferred.length > 0) {
+    return prioritizeLanguages(baseLanguages, explicitPreferred)
+  }
+
+  const envPreferred = normalizePreferredLanguages(env && env.preferredLanguages, baseLanguages)
+  if (envPreferred.length > 0) {
+    return prioritizeLanguages(baseLanguages, envPreferred)
+  }
+
+  const envLanguage = normalizeLanguageCode(env && (env.preferredLanguage || env.lang || env.language))
   if (envLanguage && baseLanguages.indexOf(envLanguage) !== -1) {
-    return prioritizeLanguage(baseLanguages, envLanguage)
+    return prioritizeLanguages(baseLanguages, envLanguage)
   }
 
   const detectedLanguage = detectDocumentPrimaryLanguage(state && state.src ? state.src : '', baseLanguages)
   if (detectedLanguage) {
-    return prioritizeLanguage(baseLanguages, detectedLanguage)
+    return prioritizeLanguages(baseLanguages, detectedLanguage)
   }
   return baseLanguages
 }
 const needsPreferredLanguagesResolution = (opt) => {
   if (!opt || !opt.markRegState || !Array.isArray(opt.markRegState.languages)) return false
+  if (!opt.autoCaptionDetection) return false
   if (opt.markRegState.languages.length <= 1) return false
-  if (Array.isArray(opt.preferredLanguages) && opt.preferredLanguages.length > 0) return false
   return opt.autoAltCaption === true || opt.autoTitleCaption === true
 }
 const normalizeOptionalClassName = (value) => {
@@ -190,10 +228,62 @@ const getLabelPrefixMarkerMatch = (inlineToken, markerReg) => {
   return match[0]
 }
 
-const parseImageAttrs = (raw) => {
+const splitImageAttrParts = (raw) => {
   if (raw === null || raw === undefined) return null
+  const parts = []
+  let current = ''
+  let quote = ''
+  let escaped = false
+  for (let i = 0; i < raw.length; i++) {
+    const ch = raw[i]
+    if (quote) {
+      current += ch
+      if (escaped) {
+        escaped = false
+        continue
+      }
+      if (ch === '\\') {
+        escaped = true
+        continue
+      }
+      if (ch === quote) {
+        quote = ''
+      }
+      continue
+    }
+    if (ch === '"' || ch === "'") {
+      quote = ch
+      current += ch
+      continue
+    }
+    if (ch === ' ') {
+      if (current) {
+        parts.push(current)
+        current = ''
+      }
+      continue
+    }
+    current += ch
+  }
+  if (quote) return null
+  if (current) parts.push(current)
+  return parts
+}
+
+const unquoteAttrValue = (value) => {
+  if (typeof value !== 'string' || value.length < 2) return value || ''
+  const first = value[0]
+  const last = value[value.length - 1]
+  if ((first === '"' && last === '"') || (first === "'" && last === "'")) {
+    return value.slice(1, -1).replace(/\\(["'\\])/g, '$1')
+  }
+  return value
+}
+
+const parseImageAttrs = (raw) => {
+  const parts = splitImageAttrParts(raw)
+  if (!parts || parts.length === 0) return null
   const attrs = []
-  const parts = raw.split(/ +/)
   for (let i = 0; i < parts.length; i++) {
     let entry = parts[i]
     if (!entry) continue
@@ -202,9 +292,15 @@ const parseImageAttrs = (raw) => {
     } else if (idAttrReg.test(entry)) {
       entry = entry.replace(idAttrReg, 'id=')
     }
-    const imageAttr = entry.match(attrParseReg)
-    if (!imageAttr || !imageAttr[1]) continue
-    attrs.push([imageAttr[1], imageAttr[2]])
+    const equalIndex = entry.indexOf('=')
+    if (equalIndex === -1) {
+      if (!attrNameReg.test(entry)) return null
+      attrs.push([entry, ''])
+      continue
+    }
+    const name = entry.slice(0, equalIndex)
+    if (!name || !attrNameReg.test(name)) return null
+    attrs.push([name, unquoteAttrValue(entry.slice(equalIndex + 1))])
   }
   return attrs
 }
@@ -214,7 +310,7 @@ const normalizeAutoLabelNumberSets = (value) => {
   if (!value) return normalized
   if (Array.isArray(value)) {
     for (const entry of value) {
-      if (normalized.hasOwnProperty(entry)) normalized[entry] = true
+      if (entry === 'img' || entry === 'table') normalized[entry] = true
     }
     return normalized
   }
@@ -299,6 +395,14 @@ const getImageAltText = (token) => {
 
 const getImageTitleText = (token) => getTokenAttr(token, 'title')
 
+const getFallbackStringLabelJoint = (label) => {
+  if (!label) return ''
+  if (labelTrailingJointReg.test(label)) {
+    return asciiLabelReg.test(label) ? ' ' : ''
+  }
+  return asciiLabelReg.test(label) ? '. ' : '　'
+}
+
 const buildCaptionWithFallback = (text, fallbackOption, mark, markRegState, preferredLanguages) => {
   const trimmedText = (text || '').trim()
   if (!fallbackOption) return ''
@@ -315,7 +419,19 @@ const buildCaptionWithFallback = (text, fallbackOption, mark, markRegState, pref
   if (generatedDefaults) {
     return label + (generatedDefaults.joint || '') + (generatedDefaults.space || '') + trimmedText
   }
-  return label + (asciiLabelReg.test(label) ? '. ' : '　') + trimmedText
+  return label + getFallbackStringLabelJoint(label) + trimmedText
+}
+
+const validateFallbackCaptionLabelOption = (optionName, fallbackOption, markRegState) => {
+  if (typeof fallbackOption !== 'string') return
+  const sampleCaption = buildCaptionWithFallback('caption', fallbackOption, 'img', markRegState, null)
+  const analysis = analyzeCaptionStart(sampleCaption, {
+    markRegState,
+    preferredMark: 'img',
+  })
+  if (!analysis || analysis.mark !== 'img' || analysis.kind !== 'caption') {
+    throw new Error(`${optionName} must be a string label recognized as an image caption by p7d-markdown-it-p-captions: ${fallbackOption}`)
+  }
 }
 
 const createAutoCaptionParagraph = (captionText, TokenConstructor) => {
@@ -343,12 +459,18 @@ const getCaptionInlineToken = (tokens, range, caption) => {
 }
 
 const hasClassName = (classAttr, className) => {
-  const index = classAttr.indexOf(className)
-  if (index === -1) return false
-  const end = index + className.length
-  if (index > 0 && classAttr.charCodeAt(index - 1) > 0x20) return false
-  if (end < classAttr.length && classAttr.charCodeAt(end) > 0x20) return false
-  return true
+  if (!classAttr || !className) return false
+  let index = 0
+  while (index < classAttr.length) {
+    index = classAttr.indexOf(className, index)
+    if (index === -1) return false
+    const end = index + className.length
+    const beforeBoundary = index === 0 || classAttr.charCodeAt(index - 1) <= 0x20
+    const afterBoundary = end >= classAttr.length || classAttr.charCodeAt(end) <= 0x20
+    if (beforeBoundary && afterBoundary) return true
+    index = end
+  }
+  return false
 }
 
 const hasAnyClassName = (classAttr, classNames) => {
@@ -387,29 +509,18 @@ const updateInlineTokenContent = (inlineToken, originalText, newText) => {
     inlineToken.content.slice(index + originalText.length)
 }
 
-const parseTrailingPositiveInteger = (text) => {
+const parsePositiveInteger = (text) => {
   if (typeof text !== 'string' || text.length === 0) return null
-  let end = text.length - 1
-  while (end >= 0 && text.charCodeAt(end) === 0x20) end--
-  if (end < 0) return null
-  const lastCode = text.charCodeAt(end)
-  if (lastCode < 0x30 || lastCode > 0x39) return null
-  let start = end
-  while (start >= 0) {
-    const code = text.charCodeAt(start)
-    if (code < 0x30 || code > 0x39) break
-    start--
-  }
   let value = 0
-  let digitBase = 1
-  for (let i = end; i > start; i--) {
-    value += (text.charCodeAt(i) - 0x30) * digitBase
-    digitBase *= 10
+  for (let i = 0; i < text.length; i++) {
+    const code = text.charCodeAt(i)
+    if (code < 0x30 || code > 0x39) return null
+    value = value * 10 + code - 0x30
   }
   return value
 }
 
-const ensureAutoFigureNumbering = (tokens, range, caption, figureNumberState, opt) => {
+const ensureAutoFigureNumbering = (tokens, range, caption, figureNumberState, sp, opt) => {
   const captionType = caption.name === 'img' ? 'img' : (caption.name === 'table' ? 'table' : '')
   if (!captionType) return
   if (!shouldApplyLabelNumbering(captionType, opt)) return
@@ -418,19 +529,12 @@ const ensureAutoFigureNumbering = (tokens, range, caption, figureNumberState, op
   const labelTextToken = getInlineLabelTextToken(inlineToken, captionType, opt)
   if (!labelTextToken || typeof labelTextToken.content !== 'string') return
   const originalText = labelTextToken.content
-  let end = originalText.length - 1
-  while (end >= 0 && originalText.charCodeAt(end) === 0x20) end--
-  if (end >= 0) {
-    const code = originalText.charCodeAt(end)
-    if (code >= 0x30 && code <= 0x39) {
-      const explicitValue = parseTrailingPositiveInteger(originalText)
-      if (explicitValue !== null) {
-        if (explicitValue > (figureNumberState[captionType] || 0)) {
-          figureNumberState[captionType] = explicitValue
-        }
-        return
-      }
+  if (sp && sp.captionDecision && sp.captionDecision.hasExplicitNumber) {
+    const explicitValue = parsePositiveInteger(sp.captionDecision.number)
+    if (explicitValue !== null && explicitValue > (figureNumberState[captionType] || 0)) {
+      figureNumberState[captionType] = explicitValue
     }
+    return
   }
   figureNumberState[captionType] = (figureNumberState[captionType] || 0) + 1
   const baseLabel = originalText.trim()
@@ -453,89 +557,87 @@ const matchAutoCaptionText = (text, opt, preferredMark = 'img') => {
   return ''
 }
 
-const getAutoCaptionFromImage = (imageToken, opt) => {
-  if (!opt.autoCaptionDetection) return ''
-  if (!opt.autoAltCaption && !opt.autoTitleCaption && !(opt.markRegState && opt.markRegState.markReg && opt.markRegState.markReg.img)) return ''
+const getAutoCaptionFromImage = (imageToken, opt, preferredLanguages) => {
+  if (!opt.autoAltCaption && !opt.autoTitleCaption && !(opt.markRegState && opt.markRegState.markReg && opt.markRegState.markReg.img)) return null
 
   const altText = getImageAltText(imageToken)
   let caption = matchAutoCaptionText(altText, opt)
   if (caption) {
-    clearImageAltAttr(imageToken)
-    return caption
+    return { text: caption, source: 'alt' }
   }
-  if (!caption && opt.autoAltCaption) {
+  if (opt.autoAltCaption) {
     const altForFallback = altText || ''
-    const fallbackCaption = buildCaptionWithFallback(altForFallback, opt.autoAltCaption, 'img', opt.markRegState, opt.preferredLanguages)
-    if (fallbackCaption && imageToken) {
-      clearImageAltAttr(imageToken)
-    }
+    const fallbackCaption = buildCaptionWithFallback(altForFallback, opt.autoAltCaption, 'img', opt.markRegState, preferredLanguages)
     caption = fallbackCaption
   }
-  if (caption) return caption
+  if (caption) return { text: caption, source: 'alt' }
 
   const titleText = getImageTitleText(imageToken)
   caption = matchAutoCaptionText(titleText, opt)
   if (caption) {
-    clearImageTitleAttr(imageToken)
-    return caption
+    return { text: caption, source: 'title' }
   }
-  if (!caption && opt.autoTitleCaption) {
+  if (opt.autoTitleCaption) {
     const titleForFallback = titleText || ''
-    const fallbackCaption = buildCaptionWithFallback(titleForFallback, opt.autoTitleCaption, 'img', opt.markRegState, opt.preferredLanguages)
-    if (fallbackCaption && imageToken) {
-      clearImageTitleAttr(imageToken)
-    }
+    const fallbackCaption = buildCaptionWithFallback(titleForFallback, opt.autoTitleCaption, 'img', opt.markRegState, preferredLanguages)
     caption = fallbackCaption
   }
-  return caption
+  return caption ? { text: caption, source: 'title' } : null
+}
+
+const consumeAutoCaptionSource = (imageToken, autoCaption) => {
+  if (!imageToken || !autoCaption) return
+  if (autoCaption.source === 'alt') {
+    clearImageAltAttr(imageToken)
+  } else if (autoCaption.source === 'title') {
+    clearImageTitleAttr(imageToken)
+  }
 }
 
 const checkPrevCaption = (tokens, n, caption, sp, opt, captionState) => {
-  if(n < 3) return caption
-  const captionStartToken = tokens[n-3]
-  const captionInlineToken = tokens[n-2]
-  const captionEndToken = tokens[n-1]
+  if (n < 3) return
+  const captionStartToken = tokens[n - 3]
+  const captionInlineToken = tokens[n - 2]
+  const captionEndToken = tokens[n - 1]
   if (captionStartToken === undefined || captionEndToken === undefined) return
   if (captionStartToken.type !== 'paragraph_open' || captionEndToken.type !== 'paragraph_close') return
-  setCaptionParagraph(n-3, captionState, caption, null, sp, opt)
+  setCaptionParagraph(n - 3, captionState, caption, null, sp, opt)
   const captionName = sp && sp.captionDecision ? sp.captionDecision.mark : ''
-  if(!captionName) {
+  if (!captionName) {
     if (opt.labelPrefixMarkerWithoutLabelPrevReg) {
-        const markerMatch = getLabelPrefixMarkerMatch(captionInlineToken, opt.labelPrefixMarkerWithoutLabelPrevReg)
-        if (markerMatch) {
-          stripLabelPrefixMarker(captionInlineToken, markerMatch)
-          caption.isPrev = true
-        }
+      const markerMatch = getLabelPrefixMarkerMatch(captionInlineToken, opt.labelPrefixMarkerWithoutLabelPrevReg)
+      if (markerMatch) {
+        stripLabelPrefixMarker(captionInlineToken, markerMatch)
+        caption.isPrev = true
+      }
     }
     return
   }
   caption.name = captionName
   caption.isPrev = true
-  return
 }
 
 const checkNextCaption = (tokens, en, caption, sp, opt, captionState) => {
-  if (en + 2 > tokens.length) return
-  const captionStartToken = tokens[en+1]
-  const captionInlineToken = tokens[en+2]
-  const captionEndToken = tokens[en+3]
+  if (en + 3 >= tokens.length) return
+  const captionStartToken = tokens[en + 1]
+  const captionInlineToken = tokens[en + 2]
+  const captionEndToken = tokens[en + 3]
   if (captionStartToken === undefined || captionEndToken === undefined) return
   if (captionStartToken.type !== 'paragraph_open' || captionEndToken.type !== 'paragraph_close') return
-  setCaptionParagraph(en+1, captionState, caption, null, sp, opt)
+  setCaptionParagraph(en + 1, captionState, caption, null, sp, opt)
   const captionName = sp && sp.captionDecision ? sp.captionDecision.mark : ''
-  if(!captionName) {
+  if (!captionName) {
     if (opt.labelPrefixMarkerWithoutLabelNextReg) {
-        const markerMatch = getLabelPrefixMarkerMatch(captionInlineToken, opt.labelPrefixMarkerWithoutLabelNextReg)
-        if (markerMatch) {
-          stripLabelPrefixMarker(captionInlineToken, markerMatch)
-          caption.isNext = true
-        }
+      const markerMatch = getLabelPrefixMarkerMatch(captionInlineToken, opt.labelPrefixMarkerWithoutLabelNextReg)
+      if (markerMatch) {
+        stripLabelPrefixMarker(captionInlineToken, markerMatch)
+        caption.isNext = true
+      }
     }
     return
   }
   caption.name = captionName
   caption.isNext = true
-  return
 }
 
 const cleanCaptionTokenAttrs = (token, captionName, opt) => {
@@ -598,59 +700,115 @@ const changePrevCaptionPosition = (tokens, n, caption, opt) => {
   const captionStartToken = tokens[n-3]
   const captionInlineToken = tokens[n-2]
   const captionEndToken = tokens[n-1]
+  const figureBaseLevel = getTokenLevel(tokens[n])
 
   cleanCaptionTokenAttrs(captionStartToken, caption.name, opt)
   captionStartToken.type = 'figcaption_open'
   captionStartToken.tag = 'figcaption'
+  captionStartToken.block = true
+  captionStartToken.level = figureBaseLevel + 1
+  captionInlineToken.level = figureBaseLevel + 2
   captionEndToken.type = 'figcaption_close'
   captionEndToken.tag = 'figcaption'
+  captionEndToken.block = true
+  captionEndToken.level = figureBaseLevel + 1
   tokens.splice(n + 2, 0, captionStartToken, captionInlineToken, captionEndToken)
   tokens.splice(n-3, 3)
-  return true
 }
 
 const changeNextCaptionPosition = (tokens, en, caption, opt) => {
-  const captionStartToken = tokens[en+2] // +1: text node for figure.
-  const captionInlineToken = tokens[en+3]
-  const captionEndToken = tokens[en+4]
+  const captionStartToken = tokens[en+1]
+  const captionInlineToken = tokens[en+2]
+  const captionEndToken = tokens[en+3]
+  const figureBaseLevel = getTokenLevel(tokens[en])
   cleanCaptionTokenAttrs(captionStartToken, caption.name, opt)
   captionStartToken.type = 'figcaption_open'
   captionStartToken.tag = 'figcaption'
+  captionStartToken.block = true
+  captionStartToken.level = figureBaseLevel + 1
+  captionInlineToken.level = figureBaseLevel + 2
   captionEndToken.type = 'figcaption_close'
   captionEndToken.tag = 'figcaption'
+  captionEndToken.block = true
+  captionEndToken.level = figureBaseLevel + 1
   tokens.splice(en, 0, captionStartToken, captionInlineToken, captionEndToken)
-  tokens.splice(en+5, 3)
-  return true
+  tokens.splice(en+4, 3)
+}
+
+const getTokenMap = (token) => {
+  return token && Array.isArray(token.map) && token.map.length === 2 ? token.map : null
+}
+
+const findNearestMapInRange = (tokens, start, end, step) => {
+  let i = start
+  while (step > 0 ? i <= end : i >= end) {
+    const map = getTokenMap(tokens[i])
+    if (map) return map
+    i += step
+  }
+  return null
+}
+
+const getRangeMap = (tokens, start, end) => {
+  const startMap = getTokenMap(tokens[start]) || findNearestMapInRange(tokens, start, end, 1)
+  if (!startMap) return null
+  const endMap = getTokenMap(tokens[end]) || findNearestMapInRange(tokens, end, start, -1) || startMap
+  const startLine = startMap[0]
+  const endLine = Math.max(startMap[1], endMap[1])
+  if (typeof startLine !== 'number' || typeof endLine !== 'number' || endLine < startLine) {
+    return [startMap[0], startMap[1]]
+  }
+  return [startLine, endLine]
+}
+
+const getTokenLevel = (token, fallback = 0) => {
+  return token && typeof token.level === 'number' ? token.level : fallback
+}
+
+const adjustTokenLevels = (tokens, start, end, delta) => {
+  if (!delta) return
+  for (let i = start; i <= end; i++) {
+    const token = tokens[i]
+    if (token && typeof token.level === 'number') {
+      token.level += delta
+    }
+  }
 }
 
 const wrapWithFigure = (tokens, range, checkTokenTagName, caption, replaceInsteadOfWrap, sp, opt, TokenConstructor) => {
   let n = range.start
   let en = range.end
+  const baseLevel = getTokenLevel(tokens[n])
+  const childLevel = baseLevel + 1
   const figureStartToken = new TokenConstructor('figure_open', 'figure', 1)
   const figureClassName = sp.figureClassName || resolveFigureClassName(checkTokenTagName, sp, opt)
   figureStartToken.attrSet('class', figureClassName)
+  figureStartToken.block = true
+  figureStartToken.level = baseLevel
 
   if (opt.roleDocExample && (checkTokenTagName === 'pre-code' || checkTokenTagName === 'pre-samp')) {
     figureStartToken.attrSet('role', 'doc-example')
   }
   const figureEndToken = new TokenConstructor('figure_close', 'figure', -1)
-  const rangeStartMap = tokens[n] && Array.isArray(tokens[n].map) && tokens[n].map.length === 2
-    ? tokens[n].map
-    : null
-  const rangeEndMap = tokens[en] && Array.isArray(tokens[en].map) && tokens[en].map.length === 2
-    ? tokens[en].map
-    : rangeStartMap
-  if (rangeStartMap) {
-    figureStartToken.map = [rangeStartMap[0], rangeStartMap[1]]
-  }
-  if (rangeEndMap) {
-    figureEndToken.map = [rangeEndMap[0], rangeEndMap[1]]
+  figureEndToken.block = true
+  figureEndToken.level = baseLevel
+  const rangeMap = getRangeMap(tokens, n, en)
+  if (rangeMap) {
+    figureStartToken.map = [rangeMap[0], rangeMap[1]]
+    figureEndToken.map = [rangeMap[0], rangeMap[1]]
   }
   const createBreakToken = () => {
     const breakToken = new TokenConstructor('text', '', 0)
     breakToken.content = '\n'
+    breakToken.level = childLevel
     return breakToken
   }
+  const createEmptyTextToken = () => {
+    const emptyToken = new TokenConstructor('text', '', 0)
+    emptyToken.content = ''
+    emptyToken.level = childLevel
+    return emptyToken
+  }
   if (caption.name === 'img') {
     const joinAttrs = (attrs) => {
       if (!attrs || attrs.length === 0) return
@@ -665,24 +823,23 @@ const wrapWithFigure = (tokens, range, checkTokenTagName, caption, replaceInstea
     joinAttrs(tokens[n].attrs)
   }
   if (replaceInsteadOfWrap) {
-    tokens.splice(en, 1, createBreakToken(), figureEndToken, createBreakToken())
-    tokens.splice(n, 1, figureStartToken, createBreakToken())
+    tokens.splice(en, 1, createBreakToken(), figureEndToken)
+    tokens.splice(n, 1, figureStartToken, createEmptyTextToken())
     en = en + 2
   } else {
-    tokens.splice(en+1, 0, figureEndToken, createBreakToken())
-    tokens.splice(n, 0, figureStartToken, createBreakToken())
+    adjustTokenLevels(tokens, n, en, 1)
+    tokens.splice(en+1, 0, figureEndToken)
+    tokens.splice(n, 0, figureStartToken, createEmptyTextToken())
     en = en + 3
   }
   range.start = n
   range.end = en
-  return
 }
 
 const checkCaption = (tokens, n, en, caption, sp, opt, captionState) => {
   checkPrevCaption(tokens, n, caption, sp, opt, captionState)
   if (caption.isPrev) return
   checkNextCaption(tokens, en, caption, sp, opt, captionState)
-  return
 }
 
 const getNestedContainerType = (token) => {
@@ -786,7 +943,7 @@ const hasLeadingImageChild = (token) => {
 const detectImageParagraph = (nextToken, n, caption, sp, opt) => {
   const multipleImagesEnabled = !!opt.multipleImages
   const styleProcessEnabled = !!opt.styleProcess
-  const allowSingleImageWithoutCaption = !!opt.oneImageWithoutCaption
+  const allowImageParagraphWithoutCaption = !!opt.imageOnlyParagraphWithoutCaption
   const children = nextToken.children
   const imageToken = children[0]
   const childrenLength = children.length
@@ -794,6 +951,7 @@ const detectImageParagraph = (nextToken, n, caption, sp, opt) => {
   let isMultipleImagesHorizontal = true
   let isMultipleImagesVertical = true
   let isValid = true
+  let trailingAttrToken = null
   caption.name = 'img'
   if (childrenLength === 1) {
     return {
@@ -801,7 +959,7 @@ const detectImageParagraph = (nextToken, n, caption, sp, opt) => {
       tagName: 'img',
       en: n + 2,
       replaceInsteadOfWrap: true,
-      wrapWithoutCaption: allowSingleImageWithoutCaption,
+      wrapWithoutCaption: allowImageParagraphWithoutCaption,
       canWrap: true,
       imageToken,
     }
@@ -825,13 +983,13 @@ const detectImageParagraph = (nextToken, n, caption, sp, opt) => {
         const imageAttrs = rawContent.match(imageAttrsReg)
         if (imageAttrs) {
           const parsedAttrs = parseImageAttrs(imageAttrs[1])
-          if (parsedAttrs && parsedAttrs.length) {
+          if (parsedAttrs) {
             for (let i = 0; i < parsedAttrs.length; i++) {
               sp.attrs.push(parsedAttrs[i])
             }
-            child.content = ''
+            trailingAttrToken = child
+            break
           }
-          break
         }
       }
       if (typeof rawContent === 'string' && rawContent.trim()) {
@@ -867,12 +1025,6 @@ const detectImageParagraph = (nextToken, n, caption, sp, opt) => {
     } else {
       caption.nameSuffix = '-multiple'
     }
-    for (let i = 0; i < childrenLength; i++) {
-      const child = children[i]
-      if (isOnlySpacesText(child)) {
-        child.content = ''
-      }
-    }
   }
   const en = n + 2
   let tagName = 'img'
@@ -882,9 +1034,25 @@ const detectImageParagraph = (nextToken, n, caption, sp, opt) => {
     tagName,
     en,
     replaceInsteadOfWrap: true,
-    wrapWithoutCaption: isValid && allowSingleImageWithoutCaption,
+    wrapWithoutCaption: isValid && allowImageParagraphWithoutCaption,
     canWrap: isValid,
     imageToken,
+    inlineToken: nextToken,
+    trailingAttrToken,
+  }
+}
+
+const applyImageParagraphTransform = (detection) => {
+  if (!detection || detection.type !== 'image') return
+  if (detection.trailingAttrToken) {
+    detection.trailingAttrToken.content = ''
+  }
+  if (!detection.inlineToken || detection.tagName === 'img') return
+  const children = detection.inlineToken.children
+  for (let i = 0; i < children.length; i++) {
+    if (isOnlySpacesText(children[i])) {
+      children[i].content = ''
+    }
   }
 }
 
@@ -894,16 +1062,19 @@ const figureWithCaption = (state, opt) => {
     table: 0,
   }
 
-  const captionState = { tokens: state.tokens, Token: state.Token }
+  const captionState = {
+    tokens: state.tokens,
+    Token: state.Token,
+    preferredLanguages: opt.preferredLanguages,
+  }
   const shouldResolvePreferredLanguages = !!(
     opt.shouldResolvePreferredLanguages &&
     sourceMayNeedPreferredLanguages(state)
   )
-  const renderOpt = shouldResolvePreferredLanguages ? Object.create(opt) : opt
   if (shouldResolvePreferredLanguages) {
-    renderOpt.preferredLanguages = resolvePreferredLanguagesForState(state, opt)
+    captionState.preferredLanguages = resolvePreferredLanguagesForState(state, opt)
   }
-  figureWithCaptionCore(state.tokens, renderOpt, figureNumberState, state.Token, captionState, null, 0)
+  figureWithCaptionCore(state.tokens, opt, figureNumberState, state.Token, captionState, null, 0)
 }
 
 const figureWithCaptionCore = (tokens, opt, figureNumberState, TokenConstructor, captionState, parentType = null, startIndex = 0) => {
@@ -984,9 +1155,9 @@ const figureWithCaptionCore = (tokens, opt, figureNumberState, TokenConstructor,
     applyCaptionDrivenFigureClass(rCaption, rSp, opt)
 
     let hasCaption = rCaption.isPrev || rCaption.isNext
-    let pendingAutoCaption = ''
+    let pendingAutoCaption = null
     if (!hasCaption && detection.type === 'image' && opt.autoCaptionDetection) {
-      pendingAutoCaption = getAutoCaptionFromImage(detection.imageToken, opt)
+      pendingAutoCaption = getAutoCaptionFromImage(detection.imageToken, opt, captionState.preferredLanguages)
       if (pendingAutoCaption) {
         hasCaption = true
       }
@@ -1002,21 +1173,18 @@ const figureWithCaptionCore = (tokens, opt, figureNumberState, TokenConstructor,
       continue
     }
 
-    let shouldWrap = false
-    if (detection.type === 'html') {
-      shouldWrap = detection.canWrap !== false && (hasCaption || detection.wrapWithoutCaption)
-    } else if (detection.type === 'image') {
-      shouldWrap = detection.canWrap !== false && (hasCaption || detection.wrapWithoutCaption)
-      if (token.hidden === true) {
-        shouldWrap = false
-      }
-    } else {
-      shouldWrap = detection.canWrap !== false && hasCaption
+    let shouldWrap = hasCaption
+    if (detection.type === 'html' || detection.type === 'image') {
+      shouldWrap = shouldWrap || detection.wrapWithoutCaption
+    }
+    if (detection.type === 'image' && token.hidden === true) {
+      shouldWrap = false
     }
 
     if (shouldWrap) {
       if (pendingAutoCaption) {
-        const captionTokens = createAutoCaptionParagraph(pendingAutoCaption, TokenConstructor)
+        consumeAutoCaptionSource(detection.imageToken, pendingAutoCaption)
+        const captionTokens = createAutoCaptionParagraph(pendingAutoCaption.text, TokenConstructor)
         tokens.splice(rRange.start, 0, ...captionTokens)
         const insertedLength = captionTokens.length
         rRange.start += insertedLength
@@ -1025,7 +1193,8 @@ const figureWithCaptionCore = (tokens, opt, figureNumberState, TokenConstructor,
         checkCaption(tokens, rRange.start, rRange.end, rCaption, rSp, opt, captionState)
         applyCaptionDrivenFigureClass(rCaption, rSp, opt)
       }
-      ensureAutoFigureNumbering(tokens, rRange, rCaption, figureNumberState, opt)
+      applyImageParagraphTransform(detection)
+      ensureAutoFigureNumbering(tokens, rRange, rCaption, figureNumberState, rSp, opt)
       wrapWithFigure(tokens, rRange, detection.tagName, rCaption, detection.replaceInsteadOfWrap, rSp, opt, TokenConstructor)
     }
 
@@ -1044,8 +1213,6 @@ const figureWithCaptionCore = (tokens, opt, figureNumberState, TokenConstructor,
       } else if (rCaption.isNext) {
         changeNextCaptionPosition(tokens, en, rCaption, opt)
         nextIndex = en + 4
-      } else {
-        nextIndex = en + 1
       }
     }
 
@@ -1061,17 +1228,18 @@ const figureWithCaptionCore = (tokens, opt, figureNumberState, TokenConstructor,
 }
 
 const mditFigureWithPCaption = (md, option) => {
-  let opt = {
+  const opt = {
     // Caption languages delegated to p-captions.
     languages: ['en', 'ja'],
-    preferredLanguages: null, // optional tie-break order for generated fallback labels; defaults to inferred document order / languages
+    preferredLanguages: null, // compatibility tie-break for generated fallback labels; prefer env.locale / env.preferredLocales per render
 
     // --- figure-wrapper behavior ---
     classPrefix: 'f',
     figureClassThatWrapsIframeTypeBlockquote: null,
     figureClassThatWrapsSlides: null,
     styleProcess: true,
-    oneImageWithoutCaption: false,
+    imageOnlyParagraphWithoutCaption: false,
+    oneImageWithoutCaption: false, // legacy alias for imageOnlyParagraphWithoutCaption
     iframeWithoutCaption: false,
     videoWithoutCaption: false,
     audioWithoutCaption: false,
@@ -1109,11 +1277,18 @@ const mditFigureWithPCaption = (md, option) => {
     labelClassFollowsFigure: false,
     figureToLabelClassMap: null,
   }
-  const hasExplicitAutoLabelNumberSets = option && Object.prototype.hasOwnProperty.call(option, 'autoLabelNumberSets')
-  const hasExplicitFigureClassThatWrapsIframeTypeBlockquote = option && Object.prototype.hasOwnProperty.call(option, 'figureClassThatWrapsIframeTypeBlockquote')
-  const hasExplicitFigureClassThatWrapsSlides = option && Object.prototype.hasOwnProperty.call(option, 'figureClassThatWrapsSlides')
-  const hasExplicitLabelClassFollowsFigure = option && Object.prototype.hasOwnProperty.call(option, 'labelClassFollowsFigure')
+  const hasExplicitAutoLabelNumberSets = hasOwnOption(option, 'autoLabelNumberSets')
+  const hasExplicitImageOnlyParagraphWithoutCaption = hasOwnOption(option, 'imageOnlyParagraphWithoutCaption')
+  const hasExplicitFigureClassThatWrapsIframeTypeBlockquote = hasOwnOption(option, 'figureClassThatWrapsIframeTypeBlockquote')
+  const hasExplicitFigureClassThatWrapsSlides = hasOwnOption(option, 'figureClassThatWrapsSlides')
+  const hasExplicitLabelClassFollowsFigure = hasOwnOption(option, 'labelClassFollowsFigure')
+  if (hasOwnOption(option, 'setFigureNumber')) {
+    throw new Error('setFigureNumber is not supported by figure-with-p-caption; use autoLabelNumber or autoLabelNumberSets')
+  }
   if (option) Object.assign(opt, option)
+  opt.imageOnlyParagraphWithoutCaption = hasExplicitImageOnlyParagraphWithoutCaption
+    ? !!opt.imageOnlyParagraphWithoutCaption
+    : !!opt.oneImageWithoutCaption
   if (!hasExplicitLabelClassFollowsFigure && opt.figureToLabelClassMap) {
     opt.labelClassFollowsFigure = true
   }
@@ -1124,6 +1299,8 @@ const mditFigureWithPCaption = (md, option) => {
   if (opt.preferredLanguages.length === 0) opt.preferredLanguages = null
   opt.normalizedOptionLanguages = normalizePreferredLanguages(opt.languages, opt.markRegState.languages)
   opt.shouldResolvePreferredLanguages = needsPreferredLanguagesResolution(opt)
+  validateFallbackCaptionLabelOption('autoAltCaption', opt.autoAltCaption, opt.markRegState)
+  validateFallbackCaptionLabelOption('autoTitleCaption', opt.autoTitleCaption, opt.markRegState)
   opt.htmlWrapWithoutCaption = {
     iframe: !!opt.iframeWithoutCaption,
     video: !!opt.videoWithoutCaption,
@@ -1171,7 +1348,7 @@ const mditFigureWithPCaption = (md, option) => {
     opt.labelPrefixMarkerWithoutLabelNextReg = null
   }
 
-  //If nextCaption has `{}` style and `f-img-multipleImages`, when upgraded to markdown-it-attrs@4.2.0, the existing script will have `{}` style on nextCaption. Therefore, since markdown-it-attrs is md.core.ruler.before('linkify'), figure_with_caption will be processed after it.
+  // Run after markdown-it-attrs has attached paragraph attrs, but before text replacements.
   md.core.ruler.before('replacements', 'figure_with_caption', (state) => {
     figureWithCaption(state, opt)
   })

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@peaceroad/markdown-it-figure-with-p-caption",
-  "version": "0.17.0",
+  "version": "0.19.0",
   "description": "A markdown-it plugin. For a paragraph with only one image, a table or code block or blockquote, and by writing a caption paragraph immediately before or after, they are converted into the figure element with the figcaption element.",
   "main": "index.js",
   "type": "module",
@@ -20,16 +20,16 @@
     "url": "https://github.com/peaceroad/p7d-markdown-it-figure-with-p-caption/issues"
   },
   "devDependencies": {
-    "@peaceroad/markdown-it-cjk-breaks-mod": "^0.1.10",
-    "@peaceroad/markdown-it-renderer-fence": "^0.6.1",
-    "@peaceroad/markdown-it-renderer-image": "^0.13.0",
-    "@peaceroad/markdown-it-strong-ja": "^0.9.0",
+    "@peaceroad/markdown-it-cjk-breaks-mod": "^0.1.14",
+    "@peaceroad/markdown-it-renderer-fence": "^0.8.0",
+    "@peaceroad/markdown-it-renderer-image": "^0.16.0",
+    "@peaceroad/markdown-it-strong-ja": "^0.9.4",
     "highlight.js": "^11.11.1",
-    "markdown-it": "^14.1.0",
-    "markdown-it-attrs": "^4.3.1"
+    "markdown-it": "^14.3.0",
+    "markdown-it-attrs": "^5.0.0"
   },
   "dependencies": {
-    "p7d-markdown-it-p-captions": "0.22.0"
+    "p7d-markdown-it-p-captions": "0.24.0"
   },
   "files": [
     "index.js",