createCodeBlockRegExp

Creates a regular expression that matches a single fenced code block anywhere in a Markdown note.

A fresh instance is returned on each call so callers may safely use stateful methods (e.g. RegExp.exec) without sharing lastIndex between scans.

Named capture groups: - LinePrefix — the leading indentation / callout blockquote prefix (> repeated once per nesting level) shared by every line of the block. - CodeBlockStartDelimiter / CodeBlockStartDelimiterChar — the opening fence (3+ backticks or tildes) and the fence character. - CodeBlockLanguage — the info-string language token immediately after the opening fence. - CodeBlockArgs — any extra arguments after the language token. - CodeBlockContent — the block body. Each content line carries the same LinePrefix. - CodeBlockEndDelimiter — the closing fence.

The CodeBlockContent group is written as \k<LinePrefix>.*(?:\n\k<LinePrefix>.*)*? — a first line followed by zero or more lines each introduced by a mandatory newline. This deliberately avoids the ambiguous (?:\n?\k<LinePrefix>.*)+? form (optional newline + repeatable line remainder), which lets a single line be partitioned in exponentially many ways and causes catastrophic backtracking — a synchronous freeze — when the scan meets a fence-open with no valid closing fence before the end of the note. The mandatory-newline form matches the same text linearly.

Import:

import { createCodeBlockRegExp } from 'obsidian-dev-utils/obsidian/code-block-reg-exp';

Signature:

function createCodeBlockRegExp(): RegExp

Returns: RegExp — A fresh global RegExp matching one fenced code block.