@asciidoctor/core 3.0.4 → 4.0.0-alpha.2

package/src/rx.js

@@ -0,0 +1,829 @@
+// ESM conversion of rx.rb
+// A collection of regular expression constants used by the parser.
+//
+// Ruby → JavaScript regex engine differences handled here:
+//
+//   Ruby \p{Alpha}  → JS \p{Alphabetic}  (Unicode Binary Property, requires 'u' flag)
+//   Ruby \p{Alnum}  → JS \p{Alphabetic}\p{N}   (inside […]) or [\p{Alphabetic}\p{N}]
+//   Ruby \p{Word}   → JS \p{Alphabetic}\p{N}\p{Pc}  (Letter + Number + Connector Punct)
+//   Ruby \p{Blank}  → JS \p{Zs}\t  (Unicode Space_Separator + tab)
+//   Ruby CC_ALL (. with /m)  → [\s\S]  (no 's' flag needed)
+//   Ruby CC_ANY (.)          → .
+//   Ruby ^ / $               → always line anchors in Ruby; in JS only with 'm' flag
+//   Ruby \A / \Z             → ^ / $ in JS (string anchors, no 'm' flag)
+//
+// IMPORTANT – 'u' flag and unset back-references:
+//   Without 'u': \n to an unset group matches the empty string (Ruby-compatible).
+//   With    'u': \n to an unset group fails (stricter).
+//   → InlineLinkRx is intentionally kept WITHOUT the 'u' flag because it relies on
+//     the (?!\2) trick (negative lookahead of an unset back-reference) to guard the
+//     angle-bracket branch.  All other patterns use 'u'.
+
+// ── Character class string constants ─────────────────────────────────────────
+// CC_* → raw content for insertion INSIDE a character class: [${CC_WORD}]
+// CG_* → complete character class GROUP for standalone use:  ${CG_WORD}
+//
+// These are runtime strings whose value contains real regex syntax (single
+// backslashes) so that String.raw`…${CC_WORD}…` produces correct regex source.
+
+export const CC_ALL = '[\\s\\S]' // any char including newlines (Ruby . with /m flag)
+export const CC_ANY = '.' // any char except newlines
+export const CC_EOL = '$' // end of line / string
+
+// \p{Alphabetic} ≈ Ruby \p{Alpha} – all Unicode alphabetic characters
+export const CC_ALPHA = '\\p{Alphabetic}' // inside [...]
+export const CG_ALPHA = '\\p{Alphabetic}' // standalone (unary property, no brackets needed)
+
+// \p{Alphabetic}\p{N} ≈ Ruby \p{Alnum} – alphabetics + all Unicode numbers
+export const CC_ALNUM = '\\p{Alphabetic}\\p{N}' // inside [...]
+export const CG_ALNUM = '[\\p{Alphabetic}\\p{N}]' // standalone group
+
+// \p{Alphabetic}\p{N}\p{Pc} ≈ Ruby \p{Word}
+// Letter + Number + Connector Punctuation (underscore, undertie, …)
+export const CC_WORD = '\\p{Alphabetic}\\p{N}\\p{Pc}' // inside [...]
+export const CG_WORD = '[\\p{Alphabetic}\\p{N}\\p{Pc}]' // standalone group
+
+// \p{Zs}\t ≈ Ruby \p{Blank} – Unicode Space_Separator category + tab
+export const CG_BLANK = '[\\p{Zs}\\t]' // standalone group
+
+// Attribute list pattern fragment: \[([^\[\]]+)\]
+// Ruby: QuoteAttributeListRxt = %(\\[([^\\[\\]]+)\\])
+export const QuoteAttributeListRxt = '\\[([^\\[\\]]+)\\]'
+
+// ── Helpers ───────────────────────────────────────────────────────────────────
+
+/**
+ * Build a regex with the Unicode flag ('u'), enabling \p{…} property escapes.
+ * @param {string} src        - Regex source string (use String.raw for easy authoring).
+ * @param {string} extraFlags - Additional flags, e.g. 'm' for multiline ^ / $
+ */
+const ru = (src, extraFlags = '') => new RegExp(src, `u${extraFlags}`)
+
+/**
+ * Escape all regex metacharacters in str (equivalent to Regexp.escape in Ruby).
+ */
+function escapeRegex(str) {
+  return str.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')
+}
+
+/**
+ * Build a lazy-initialised regex map, mirroring Ruby's Hash.new { |h,k| h[k] = … }.
+ * Accessing map[key] creates and caches the regex for that key.
+ */
+function makeLazyRxMap(buildFn) {
+  const cache = new Map()
+  return new Proxy(Object.create(null), {
+    get(_target, key) {
+      if (typeof key !== 'string') return undefined
+      if (!cache.has(key)) cache.set(key, buildFn(key))
+      return cache.get(key)
+    },
+  })
+}
+
+// ── Document header ───────────────────────────────────────────────────────────
+
+/**
+ * Matches the author info line immediately following the document title.
+ * @example
+ * Doc Writer <doc@example.com>
+ * Mary_Sue Brontë
+ */
+export const AuthorInfoLineRx = ru(
+  String.raw`^(${CG_WORD}[${CC_WORD}\-'.]*)(?: +(${CG_WORD}[${CC_WORD}\-'.]*))?` +
+    String.raw`(?: +(${CG_WORD}[${CC_WORD}\-'.]*))?(?:[ ]+<([^>]+)>)?$`
+)
+
+/**
+ * Matches the delimiter that separates multiple authors.
+ * @example
+ * Doc Writer; Junior Writer
+ */
+export const AuthorDelimiterRx = /;(?: |$)/
+
+/**
+ * Matches the revision info line immediately following the author info line.
+ * @example
+ * v1.0
+ * 2013-01-01
+ * v1.0, 2013-01-01: Ring in the new year release
+ */
+export const RevisionInfoLineRx =
+  /^(?:[^\d{]*(.*?),)? *(?!:)(.*?)(?: *(?!^),?: *(.*))?$/
+
+/**
+ * Matches the title and volnum in the manpage doctype.
+ * @example
+ * = asciidoctor(1)
+ * = asciidoctor ( 1 )
+ */
+export const ManpageTitleVolnumRx = /^(.+?) *\( *(.+?) *\)$/
+
+/**
+ * Matches the name and purpose in the manpage doctype.
+ * @example
+ * asciidoctor - converts AsciiDoc source files to HTML, DocBook and other formats
+ */
+export const ManpageNamePurposeRx = /^(.+?) +- +(.+)$/
+
+// ── Preprocessor directives ───────────────────────────────────────────────────
+
+/**
+ * Matches a conditional preprocessor directive (ifdef, ifndef, ifeval, endif).
+ * @example
+ * ifdef::basebackend-html[]
+ * ifeval::["{asciidoctor-version}" >= "0.1.0"]
+ * endif::[]
+ */
+export const ConditionalDirectiveRx =
+  /^(\\)?(ifdef|ifndef|ifeval|endif)::(\S*?(?:([,+])\S*?)?)\[(.+)?\]$/
+
+/**
+ * Matches a restricted (safe) eval expression.
+ * @example
+ * "{asciidoctor-version}" >= "0.1.0"
+ */
+export const EvalExpressionRx = /^(.+?) *([=!><]=|[><]) *(.+)$/
+
+/**
+ * Matches an include preprocessor directive.
+ * @example
+ * include::chapter1.ad[]
+ * include::example.txt[lines=1;2;5..10]
+ */
+export const IncludeDirectiveRx =
+  /^(\\)?include::([^\s[](?:[^[]*[^\s[])?)\[(.+)?\]$/
+
+/**
+ * Matches a trailing tag directive in an include file.
+ *
+ * NOTE: 'm' flag required so that $ matches end-of-line (not only end-of-string) in JS.
+ * NOTE: accounts for \r in Windows line endings.
+ * @example
+ * // tag::try-catch[]
+ * // end::try-catch[]
+ */
+export const TagDirectiveRx = /\b(?:tag|(e)nd)::(\S+?)\[\](?=$|[ \r])/m
+
+// ── Attribute entries and references ─────────────────────────────────────────
+
+/**
+ * Matches a document attribute entry.
+ * @example
+ * :foo: bar
+ * :First Name: Dan
+ * :sectnums!:
+ */
+export const AttributeEntryRx = ru(
+  String.raw`^:(!?${CG_WORD}[^:]*):(?:[ \t]+(.*))?$`
+)
+
+/** Matches invalid characters in an attribute name. */
+export const InvalidAttributeNameCharsRx = ru(String.raw`[^${CC_WORD}\-]`)
+
+/**
+ * Matches a pass inline macro surrounding an attribute entry value.
+ *
+ * NOTE: ^ / $ are string anchors here (no 'm' flag). [\s\S]* allows multi-line values.
+ * @example
+ * pass:[text]
+ * pass:a[{a} {b} {c}]
+ */
+export const AttributeEntryPassMacroRx =
+  /^pass:([a-z]+(?:,[a-z-]+)*)?\[([\s\S]*)\]$/
+
+/**
+ * Matches an inline attribute reference.
+ * @example
+ * {foobar}
+ * {counter:sequence-name:1}
+ * {set:foo:bar}
+ */
+export const AttributeReferenceRx = ru(
+  String.raw`(\\)?\{(${CG_WORD}[${CC_WORD}\-]*|(set|counter2?):.*?)(\\)?\}`
+)
+
+// ── Paragraphs and delimited blocks ──────────────────────────────────────────
+
+/**
+ * Matches an anchor (id + optional reference text) on a line above a block.
+ * @example
+ * [[idname]]
+ * [[idname,Reference Text]]
+ */
+export const BlockAnchorRx = ru(
+  String.raw`^\[\[(?:|([${CC_ALPHA}_:][${CC_WORD}\-:.]*)(?:, *(.+))?)\]\]$`
+)
+
+/**
+ * Matches an attribute list above a block element.
+ * @example
+ * [quote, Adam Smith, Wealth of Nations]
+ * [{lead}]
+ */
+export const BlockAttributeListRx = ru(
+  String.raw`^\[(|[${CC_WORD}.#%{,"'].*)\]$`
+)
+
+/** Combined pattern matching either a block anchor or a block attribute list. */
+export const BlockAttributeLineRx = ru(
+  String.raw`^\[(?:|[${CC_WORD}.#%{,"'].*|\[(?:|[${CC_ALPHA}_:][${CC_WORD}\-:.]*(?:, *.+)?)\])\]$`
+)
+
+/**
+ * Matches a title above a block.
+ * @example
+ * .Title goes here
+ */
+export const BlockTitleRx = /^\.(\.?[^ \t.].*)$/
+
+/**
+ * Matches an admonition label at the start of a paragraph.
+ * @example
+ * NOTE: Just a little note.
+ * TIP: Don't forget!
+ */
+export const AdmonitionParagraphRx =
+  /^(NOTE|TIP|IMPORTANT|WARNING|CAUTION):[ \t]+/
+
+/**
+ * Matches a literal paragraph (line preceded by at least one space or tab).
+ * @example
+ * <SPACE>Foo
+ * <TAB>Foo
+ */
+export const LiteralParagraphRx = /^([ \t]+.*)$/
+
+// ── Section titles ────────────────────────────────────────────────────────────
+
+/**
+ * Matches an Atx (single-line) section title.
+ * @example
+ * == Foo
+ * == Foo ==
+ */
+export const AtxSectionTitleRx = /^(=={0,5})[ \t]+(.+?)(?:[ \t]+\1)?$/
+
+/** Extended Atx section title supporting the Markdown variant (#). */
+export const ExtAtxSectionTitleRx =
+  /^(=={0,5}|##{0,5})[ \t]+(.+?)(?:[ \t]+\1)?$/
+
+/**
+ * Matches the first line of a Setext (two-line) section title.
+ * Must not start with '.' and must contain at least one alphanumeric character.
+ */
+export const SetextSectionTitleRx = ru(String.raw`^((?!\.).*?${CG_ALNUM}.*)$`)
+
+/**
+ * Matches an anchor inside a section title.
+ * @example
+ * Section Title [[idname]]
+ * Section Title [[idname,Reference Text]]
+ */
+export const InlineSectionAnchorRx = ru(
+  String.raw` (\\)?\[\[([${CC_ALPHA}_:][${CC_WORD}\-:.]*)(?:, *(.+?))?\]\]$`
+)
+
+/**
+ * Matches invalid ID characters in a section title.
+ * NOTE: Uppercase excluded; expression is run only on a lowercase string.
+ */
+export const InvalidSectionIdCharsRx = ru(
+  String.raw`<[^>]+>|&(?:[a-z][a-z]+\d{0,2}|#\d\d\d{0,4}|#x[\da-f][\da-f][\da-f]{0,3});|[^ ${CC_WORD}\-.]+?`
+)
+
+/** Matches an explicit section level style like sect1. */
+export const SectionLevelStyleRx = /^sect\d$/
+
+// ── Lists ─────────────────────────────────────────────────────────────────────
+
+/**
+ * Detects the start of any list item.
+ *
+ * NOTE: Check only up to the blank character since non-whitespace follows.
+ * IMPORTANT: Must agree with the per-list-type regexps or the parser will hang.
+ */
+export const AnyListRx =
+  /^(?:[ \t]*(?:-|\*\**|\.\.*|\u2022|\d+\.|[a-zA-Z]\.|[IVXivx]+\))[ \t]|(?!\/\/[^/])[ \t]*[^ \t].*?(?::::{0,2}|;;)(?:$|[ \t])|<(?:\d+|\.)>[ \t])/
+
+/**
+ * Matches an unordered list item.
+ * @example
+ * * Foo
+ * - Foo
+ */
+export const UnorderedListRx = /^[ \t]*(-|\*\**|\u2022)[ \t]+([\s\S]*)$/
+
+/**
+ * Matches an ordered list item.
+ * @example
+ * . Foo    1. Foo    a. Foo    I. Foo
+ */
+export const OrderedListRx =
+  /^[ \t]*(\.\.*|\d+\.|[a-zA-Z]\.|[IVXivx]+\))[ \t]+([\s\S]*)$/
+
+/** Ordinal pattern for each ordered list type. */
+export const OrderedListMarkerRxMap = {
+  arabic: /\d+\./,
+  loweralpha: /[a-z]\./,
+  lowerroman: /[ivx]+\)/,
+  upperalpha: /[A-Z]\./,
+  upperroman: /[IVX]+\)/,
+}
+
+/**
+ * Matches a description list entry.
+ * @example
+ * foo::
+ * foo:: The metasyntactic variable …
+ */
+export const DescriptionListRx =
+  /^(?!\/\/[^/])[ \t]*([^ \t].*?)(:::{0,2}|;;)(?:$|[ \t]+([\s\S]*)$)/
+
+/** Matches a sibling description list item (excluding the delimiter given by key). */
+export const DescriptionListSiblingRx = {
+  '::': /^(?!\/\/[^/])[ \t]*([^ \t].*?[^:]|[^ \t:])(::)(?:$|[ \t]+([\s\S]*)$)/,
+  ':::':
+    /^(?!\/\/[^/])[ \t]*([^ \t].*?[^:]|[^ \t:])(:::)(?:$|[ \t]+([\s\S]*)$)/,
+  '::::':
+    /^(?!\/\/[^/])[ \t]*([^ \t].*?[^:]|[^ \t:])(::::)(?:$|[ \t]+([\s\S]*)$)/,
+  ';;': /^(?!\/\/[^/])[ \t]*([^ \t].*?)(;;)(?:$|[ \t]+([\s\S]*)$)/,
+}
+
+/**
+ * Matches a callout list item.
+ * @example
+ * <1> Explanation
+ * <.> Explanation with automatic number
+ */
+export const CalloutListRx = /^<(\d+|\.)>[ \t]+([\s\S]*)$/
+
+/**
+ * Matches a callout reference inside literal text (applied line-by-line).
+ *
+ * Group layout:
+ *   1 – optional line-comment prefix (//  #  --  ;;)
+ *   2 – backslash escape
+ *   3 – optional XML comment delimiter (--)
+ *   4 – callout number or dot
+ */
+export const CalloutExtractRx =
+  /((?:\/\/|#|--|;;) ?)?(\\)?<!?(|--)(\d+|\.)\3>(?=(?: ?\\?<!?\3(?:\d+|\.)\3>)*$)/m
+
+/**
+ * Template string for CalloutExtractRxMap entries.
+ * Runtime value: (\\)?<()(\d+|\.)>(?=(?: ?\\?<(?:\d+|\.)>)*$)
+ * Note: 'm' flag added so $ matches end-of-line (Ruby regex default behaviour).
+ */
+export const CalloutExtractRxt =
+  '(\\\\)?<()([\\d]+|\\.)>(?=(?: ?\\\\?<(?:\\d+|\\.)>)*$)'
+
+/**
+ * Lazy map: line-comment string → callout-extract regex.
+ * Mirrors Ruby: Hash.new { |h,k| h[k] = /(prefix)?#{CalloutExtractRxt}/ }
+ */
+export const CalloutExtractRxMap = makeLazyRxMap((key) => {
+  const prefix = key ? `(${escapeRegex(key)} ?)?` : '()?'
+  return new RegExp(`${prefix}${CalloutExtractRxt}`, 'm')
+})
+
+/** Matches a callout reference when scanning source (special chars NOT yet replaced). */
+export const CalloutScanRx =
+  /\\?<!?(|--)(\d+|\.)\1>(?=(?: ?\\?<!?\1(?:\d+|\.)\1>)*$)/m
+
+/**
+ * Matches a callout reference in HTML output (special chars already replaced).
+ *
+ * Group layout mirrors CalloutExtractRx.
+ * Note: 'm' flag so $ matches end-of-line, matching Ruby regex semantics.
+ */
+export const CalloutSourceRx =
+  /((?:\/\/|#|--|;;) ?)?(\\)?&lt;!?(|--)(\d+|\.)\3&gt;(?=(?: ?\\?&lt;!?\3(?:\d+|\.)\3&gt;)*$)/m
+
+/**
+ * Template string for CalloutSourceRxMap entries.
+ * Runtime value: (\\)?&lt;()(\d+|\.)&gt;(?=(?: ?\\?&lt;(?:\d+|\.)&gt;)*$)
+ */
+export const CalloutSourceRxt =
+  '(\\\\)?&lt;()([\\d]+|\\.)&gt;(?=(?: ?\\\\?&lt;(?:\\d+|\\.)&gt;)*$)'
+
+/** Lazy map: line-comment string → callout-source regex. */
+export const CalloutSourceRxMap = makeLazyRxMap((key) => {
+  const prefix = key ? `(${escapeRegex(key)} ?)?` : '()?'
+  return new RegExp(`${prefix}${CalloutSourceRxt}`, 'm')
+})
+
+/** Dynamic map from list context to its regex. */
+export const ListRxMap = {
+  ulist: UnorderedListRx,
+  olist: OrderedListRx,
+  dlist: DescriptionListRx,
+  colist: CalloutListRx,
+}
+
+// ── Tables ────────────────────────────────────────────────────────────────────
+
+/**
+ * Parses the column test (colspec) for a table.
+ * @example
+ * 1*h,2*,^3e
+ */
+export const ColumnSpecRx =
+  /^(?:(\d+)\*)?([<^>](?:\.[<^>]?)?|(?:[<^>]?\.)?[<^>])?(\d+%?|~)?([a-z])?$/
+
+/**
+ * Parses the start of a cell test.
+ * @example
+ * 2.3+<.>m
+ */
+export const CellSpecStartRx =
+  /^[ \t]*(?:(\d+(?:\.\d*)?|(?:\d*\.)?\d+)([*+]))?([<^>](?:\.[<^>]?)?|(?:[<^>]?\.)?[<^>])?([a-z])?$/
+
+/** Parses the end of a cell test. */
+export const CellSpecEndRx =
+  /[ \t]+(?:(\d+(?:\.\d*)?|(?:\d*\.)?\d+)([*+]))?([<^>](?:\.[<^>]?)?|(?:[<^>]?\.)?[<^>])?([a-z])?$/
+
+// ── Block macros ──────────────────────────────────────────────────────────────
+
+/**
+ * Matches the custom block macro pattern.
+ * @example
+ * gist::123456[]
+ */
+export const CustomBlockMacroRx = ru(
+  String.raw`^(${CG_WORD}[${CC_WORD}\-]*)::(|\S|\S.*?\S)\[(.+)?\]$`
+)
+
+/**
+ * Matches an image, video or audio block macro.
+ * @example
+ * image::filename.png[Caption]
+ * video::http://youtube.com/12345[Cats vs Dogs]
+ */
+export const BlockMediaMacroRx = /^(image|video|audio)::(\S|\S.*?\S)\[(.+)?\]$/
+
+/**
+ * Matches the TOC block macro.
+ * @example
+ * toc::[]
+ * toc::[levels=2]
+ */
+export const BlockTocMacroRx = /^toc::\[(.+)?\]$/
+
+// ── Inline macros ─────────────────────────────────────────────────────────────
+
+/**
+ * Matches an anchor (id + optional reference text) in the flow of text.
+ *
+ * Group layout:
+ *   1 – backslash escape
+ *   2 – id  (double-bracket form)
+ *   3 – reftext  (double-bracket form)
+ *   4 – id  (anchor: macro form)
+ *   5 – reftext  (anchor: macro form)
+ * @example
+ * [[idname]]
+ * [[idname,Reference Text]]
+ * anchor:idname[]
+ * anchor:idname[Reference Text]
+ */
+export const InlineAnchorRx = ru(
+  String.raw`(\\)?(?:\[\[([${CC_ALPHA}_:][${CC_WORD}\-:.]*)(?:, *(.+?))? ?\]\]` +
+    String.raw`|anchor:([${CC_ALPHA}_:][${CC_WORD}\-:.]*)\[(?:\]|([\s\S]*?[^\\])\]))`
+)
+
+/** Scans for a non-escaped anchor in the flow of text. */
+export const InlineAnchorScanRx = ru(
+  String.raw`(?:^|[^\\\[])\[\[([${CC_ALPHA}_:][${CC_WORD}\-:.]*)(?:, *(.+?))? ?\]\]` +
+    String.raw`|(?:^|[^\\])anchor:([${CC_ALPHA}_:][${CC_WORD}\-:.]*)\[(?:\]|(.*?[^\\])\])`
+)
+
+/** Scans for a leading, non-escaped anchor. */
+export const LeadingInlineAnchorRx = ru(
+  String.raw`^\[\[([${CC_ALPHA}_:][${CC_WORD}\-:.]*)(?:, *(.+?))?\]\]`
+)
+
+/**
+ * Matches a bibliography anchor at the start of a list item.
+ * @example
+ * [[[Fowler_1997]]] Fowler M. ...
+ */
+export const InlineBiblioAnchorRx = ru(
+  String.raw`^\[\[\[([${CC_ALPHA}_:][${CC_WORD}\-:.]*)(?:, *(.+?))?\]\]\]`
+)
+
+/**
+ * Matches an inline e-mail address.
+ * @example
+ * doc.writer@example.com
+ */
+export const InlineEmailRx = ru(
+  String.raw`([\\>:/])?${CG_WORD}(?:&amp;|[${CC_WORD}\-.%+])*` +
+    String.raw`@${CG_ALNUM}[${CC_ALNUM}_\-.]*\.[a-zA-Z]{2,5}\b`
+)
+
+/**
+ * Matches an inline footnote macro (may span multiple lines).
+ *
+ * NOTE: [\s\S]*? allows multiline content (Ruby /m + CC_ALL).
+ * NOTE: (?!</a>) avoids matching inside an anchor tag.
+ * @example
+ * footnote:[text]
+ * footnote:id[text]
+ * footnoteref:[id,text]  (legacy)
+ */
+export const InlineFootnoteMacroRx = ru(
+  String.raw`\\?footnote(?:(ref):|:([${CC_WORD}\-]+)?)\[(?:|([\s\S]*?[^\\]))\](?!</a>)`
+)
+
+/**
+ * Matches an image or icon inline macro (may span multiple lines).
+ * @example
+ * image:filename.png[Alt Text]
+ * icon:github[large]
+ */
+export const InlineImageMacroRx =
+  /\\?i(?:mage|con):([^:\s[](?:[^\n[]*[^\s[])?)\[(|[\s\S]*?[^\\])\]/
+
+/**
+ * Matches an indexterm inline macro (may span multiple lines).
+ * @example
+ * indexterm:[Tigers,Big cats]
+ * (((Tigers,Big cats)))
+ * ((Tigers))
+ */
+export const InlineIndextermMacroRx =
+  /\\?(?:(indexterm2?):\[([\s\S]*?[^\\])\]|\(\(([\s\S]+?)\)\)(?!\)))/
+
+/**
+ * Matches either the kbd or btn inline macro (may span multiple lines).
+ * @example
+ * kbd:[F3]     kbd:[Ctrl+Shift+T]     btn:[Save]
+ */
+export const InlineKbdBtnMacroRx = /(\\)?(kbd|btn):\[([\s\S]*?[^\\])\]/
+
+/**
+ * Matches an implicit link and the link inline macro.
+ *
+ * NOTE: This is the Opal/JS variant of the pattern.
+ *   Group 2 captures ':' inside a lookahead from the &lt;<protocol> branch.
+ *   (?!\2) then guards the &gt;-terminated branch: when group 2 IS ':',
+ *   the guard prevents matching '://' at the start of the path; when group 2
+ *   is UNSET (other prefix branches), (?!\2) expands to (?!"") which ALWAYS
+ *   FAILS – correctly preventing the &gt; branch for non-&lt; prefixes.
+ *
+ * *** NO 'u' FLAG: the (?!\2) guard relies on unset back-references matching
+ *     the empty string, which only holds in non-Unicode mode. ***
+ *
+ * Group layout:
+ *   1 – prefix (^, link:, blank, \\?&lt; or punctuation)
+ *   2 – ':' captured by lookahead  (only when prefix is \\?&lt;)
+ *   3 – URL scheme + ://
+ *   4 – target before [   (formal macro)
+ *   5 – attrlist           (formal macro, may be empty)
+ *   6 – target before &gt; (angle-bracket autolink, requires &lt; prefix)
+ *   7 – target             (bare autolink)
+ *   8 – last non-terminating char of bare target
+ * @example
+ * https://github.com
+ * https://github.com[GitHub]
+ * <https://github.com>
+ * link:https://github.com[]
+ */
+export const InlineLinkRx =
+  /(^|link:|[ \t\u00a0]|\\?&lt;(?=\\?(?:https?|file|ftp|irc)(:))|[>()[\];"'])(\\?(?:https?|file|ftp|irc):\/\/)(?:([^\s[\]]+)\[(|[\s\S]*?[^\\])\]|(?!\2)([^\s]+?)&gt;|([^\s[\]<]*([^\s,.?![\]<)])))/m
+
+/**
+ * Matches a link or e-mail inline macro (may span multiple lines).
+ * @example
+ * link:path[label]
+ * mailto:doc.writer@example.com[]
+ */
+export const InlineLinkMacroRx =
+  /\\?(?:link|(mailto)):(|[^:\s[][^\s[]*)\[(|[\s\S]*?[^\\])\]/
+
+/** Matches the name of a macro. */
+export const MacroNameRx = ru(String.raw`^${CG_WORD}[${CC_WORD}\-]*$`)
+
+/**
+ * Matches a stem (and alternatives) inline macro (may span multiple lines).
+ * @example
+ * stem:[x != 0]
+ * latexmath:[\sqrt{4} = 2]
+ */
+export const InlineStemMacroRx =
+  /\\?(stem|(?:latex|ascii)math):([a-z]+(?:,[a-z-]+)*)?\[([\s\S]*?[^\\])\]/
+
+/**
+ * Matches a menu inline macro (may span multiple lines).
+ * @example
+ * menu:File[Save As...]
+ * menu:View[Page Style > No Style]
+ */
+export const InlineMenuMacroRx = ru(
+  String.raw`\\?menu:(${CG_WORD}|[${CC_WORD}&][^\n\[]*[^\s\[])` +
+    String.raw`\[ *(?:|([\s\S]*?[^\\]))\]`
+)
+
+/**
+ * Matches an implicit menu inline macro.
+ * @example
+ * "File > New..."
+ */
+export const InlineMenuRx = ru(
+  String.raw`\\?"([${CC_WORD}&][^"]*?[ \n]+&gt;[ \n]+[^"]*)"`
+)
+
+/**
+ * Matches an inline passthrough (may span multiple lines).
+ *
+ * Group layout (false / non-compat):
+ *   1 – preceding context or escape boundary
+ *   2 – '[' captured by lookahead (back-reference trick for attribute list detection)
+ *   3 – x- / 'attrlist x-' content
+ *   4 – QuoteAttributeListRxt content
+ *   5 – optional backslash before opening delimiter
+ *   6 – full quoted span (including delimiters)
+ *   7 – opening/closing delimiter (+ or `)
+ *   8 – span content
+ *
+ * Group layout (true / compat):
+ *   1 – preceding char or start-of-line
+ *   2 – ($) end-of-string sentinel  (never matches in inline text, preserves group count)
+ *   3 – empty group paired with sentinel
+ *   4 – QuoteAttributeListRxt content
+ *   5 – optional backslash before opening delimiter
+ *   6 – full quoted span
+ *   7 – opening/closing delimiter (`)
+ *   8 – span content
+ *
+ * NOTE: 'u' flag used, but the 'm' flag is also set so that ^ is a line anchor.
+ *   Unset optional back-references (\5?) with 'u' flag: the '?' quantifier
+ *   allows 0 occurrences, so the match continues even when the group is unset.
+ * @example
+ * +text+
+ * [x-]+text+
+ * `text`  (compat only)
+ */
+export const InlinePassRx = {
+  false: [
+    '+',
+    '-]',
+    ru(
+      String.raw`((?:^|[^${CC_WORD};:\\])(?=(\[)|\+)|\\(?=\[)|(?=\\\+))` +
+        String.raw`(?:\2(x-|[^\[\]]+ x-)\]|(?:` +
+        QuoteAttributeListRxt +
+        String.raw`)?(?=(\\)?\+))` +
+        String.raw`(\5?(\+|` +
+        '`' +
+        String.raw`)(\S|\S` +
+        CC_ALL +
+        String.raw`*?\S)\7)(?!${CG_WORD})`,
+      'm'
+    ),
+  ],
+  true: [
+    '`',
+    null,
+    ru(
+      String.raw`(^|[^` +
+        '`' +
+        String.raw`${CC_WORD}])(?:($)()|(?:` +
+        QuoteAttributeListRxt +
+        String.raw`)(?=(\\?)))?` +
+        String.raw`(\5?(` +
+        '`' +
+        String.raw`)([^` +
+        '`' +
+        String.raw`\s]|[^` +
+        '`' +
+        String.raw`\s]` +
+        CC_ALL +
+        String.raw`*?\S)\7)(?![` +
+        '`' +
+        String.raw`${CC_WORD}])`,
+      'm'
+    ),
+  ],
+}
+
+/**
+ * Matches several variants of the passthrough inline macro (may span multiple lines).
+ *
+ * Group layout:
+ *   1 – optional backslash before attribute list
+ *   2 – attribute list content  (QuoteAttributeListRxt)
+ *   3 – backslash(es) before delimiter  (0–2)
+ *   4 – delimiter: +++, ++, or $$
+ *   5 – content between delimiters  (\4 closes)
+ *   6 – backslash before pass: macro
+ *   7 – subs list after pass:
+ *   8 – content inside pass:[…]
+ * @example
+ * +++text+++
+ * $$text$$
+ * pass:quotes[text]
+ * pass:[]
+ */
+export const InlinePassMacroRx = new RegExp(
+  `(?:(?:(\\\\?)${QuoteAttributeListRxt})?(\\\\{0,2})(\\+\\+\\+?|\\$\\$)([\\s\\S]*?)\\4|(\\\\?)pass:([a-z]+(?:,[a-z-]+)*)?\\[(|[\\s\\S]*?[^\\\\])\\])`
+)
+
+/**
+ * Matches an xref (cross-reference) inline macro (may span multiple lines).
+ *
+ * NOTE: { included to support targets beginning with an attribute reference.
+ * NOTE: Special characters are already entity-encoded in the matched text.
+ *
+ * Group layout:
+ *   1 – target of <<…>> form
+ *   2 – target of xref:…[] form
+ *   3 – link text inside xref:…[…]
+ * @example
+ * <<id,reftext>>
+ * xref:id[reftext]
+ */
+export const InlineXrefMacroRx = ru(
+  String.raw`\\?(?:&lt;&lt;([${CC_WORD}#/.:{]` +
+    CC_ALL +
+    String.raw`*?)&gt;&gt;` +
+    String.raw`|xref:([${CC_WORD}#/.:{]` +
+    CC_ALL +
+    String.raw`*?)\[(?:\]|(` +
+    CC_ALL +
+    String.raw`*?[^\\])\]))`
+)
+
+// ── Layout ────────────────────────────────────────────────────────────────────
+
+/**
+ * Matches a trailing + preceded by at least one space, forcing a hard line break.
+ *
+ * NOTE: 'm' flag required so that ^ / $ are line anchors (not string anchors) in JS.
+ * @example
+ * Humpty Dumpty sat on a wall, +
+ * Humpty Dumpty had a great fall.
+ */
+export const HardLineBreakRx = /^(.*) \+$/m
+
+/**
+ * Matches a Markdown horizontal rule.
+ * @example
+ * --- or - - -
+ * *** or * * *
+ * ___ or _ _ _
+ */
+export const MarkdownThematicBreakRx = /^ {0,3}([-*_])( *)\1\2\1$/
+
+/**
+ * Matches an AsciiDoc or Markdown horizontal rule, or an AsciiDoc page break.
+ * @example
+ * '''  <<<  ---  ***  ___
+ */
+export const ExtLayoutBreakRx = /^(?:'{3,}|<{3,}|([-*_])( *)\1\2\1)$/
+
+// ── General ───────────────────────────────────────────────────────────────────
+
+/** Matches consecutive blank lines. */
+export const BlankLineRx = /\n{2,}/
+
+/**
+ * Matches whitespace escaped by a backslash.
+ * @example
+ * three\ blind\ mice
+ */
+export const EscapedSpaceRx = /\\([ \t\n])/g
+
+/** Detects text that may contain replaceable characters. */
+export const ReplaceableTextRx = /[&']|--|\.\.\.|\([CRT]M?\)/
+
+/**
+ * Matches a whitespace delimiter (space, tab, newline).
+ * Replicates the parsing rules of Ruby %w strings.
+ *
+ * TODO: Replace with /(?<!\\)[ \t\n]+/ when lookbehind is universally available.
+ */
+export const SpaceDelimiterRx = /([^\\])[ \t\n]+/g
+
+/** Matches a + or - modifier in a subs list. */
+export const SubModifierSniffRx = /[+-]/
+
+/**
+ * Matches one or more consecutive digits at the end of a line.
+ * @example
+ * docbook5   html5
+ */
+export const TrailingDigitsRx = /\d+$/
+
+/**
+ * Detects strings that resemble URIs.
+ *
+ * NOTE: ^ is used as a string-start anchor (no 'm' flag), equivalent to Ruby \A.
+ * NOTE: Does NOT match Windows paths like c:/sample.adoc or c:\sample.adoc.
+ * @example
+ * http://domain    https://domain    file:///path    data:info
+ */
+export const UriSniffRx = ru(String.raw`^${CG_ALPHA}[${CC_ALNUM}.+\-]+:\/{0,2}`)
+
+/** Detects XML tags. */
+export const XmlSanitizeRx = /<[^>]+>/