Match quotes inside strings. Images generated with (https://jex.im/regulex)

Escaped Quote Matcher.md

Escaped Quote Matcher

This is a javascript function that matches quotes while disregarding all escaped quotes.

Calling

outputArray = matchQuotes(
  inputString,
  chars,
  lengths,
  internals
);

Options:

const chars = {
  open: '\\1',
  close: '"\'',
  esc: '\\'
};
const lengths = {
  min: 0,
  max: Infinity
}
// internals is very dynamic (see: matchQuotes.js line 60)
Returns an array of matches.

Internals: matchQuotes.js line 60

example1.md

Code...

Start quote: '[' End quote: ']' Escape Char: '\'

let out = matchQuotes(
  'test[key\\[sub\\]]', [
    '[', // Start Quote
    ']', // End Quote
    '\\' // Escape Quote
  ]
);

Generated regex:

/
([\]])(?!(?:[\\]{2})*[\\](?![\\]))
((?:
  (?:[\]](?=(?:[\\]{2})*[\\](?![\\])))|
  (?:[\[](?=(?:[\\]{2})*[\\](?![\\])))|
  (?:[^\[\]])
){0,})
([\[])(?!(?:[\\]{2})*[\\](?![\\]))
/g

Output

[
  "key[sub]"
]

example2.md

Code...

Start quote: '"' or "'" End quote: '"' or "'" Escape Char: '\'

let out = matchQuotes('my \\"new\\" string and \"this should \\"be\\" matched\"');

Generated regex:

/
(['"])(?!(?:[\\]{2})*[\\](?![\\]))
((?:
  (?:\1(?=(?:[\\]{2})*[\\](?![\\])))|
  (?:(?!\1).)
){0,})
(\1)(?!(?:[\\]{2})*[\\](?![\\]))
/g

Output

[
  "this should \"be\" matched"
]

example3.md

Code...

let out = matchQuotes(
	'my \\"new\\" “strin\\“g\\” and” \"this should \\"be\\" matched\"', [
		'“', // Start
		'”', // End
		'\\' // Escape
	]
);

Generated regex:

/
([”])(?!(?:[\\]{2})*[\\](?![\\]))
((?:
	(?:[”](?=(?:[\\]{2})*[\\](?![\\])))|
	(?:[“](?=(?:[\\]{2})*[\\](?![\\])))|
	(?:[^“”])
){0,})
([“])(?!(?:[\\]{2})*[\\](?![\\]))
/g

Output

[
	"strin“g” and"
]

matchQuotes.js

function matchQuotes(string, chars, lengths, internals) {
	const reversed = string.split('').reverse().join('');
	const matches = [];
	let match;
	let open, close, esc;
	let min, max;
	
	if (typeof chars === 'string')
		close = chars;
	
	else if (chars instanceof Array) {
		open =  chars[0] || undefined;
		close = chars[1] || undefined;
		esc =   chars[2] || undefined;
	}
	
	else if (chars instanceof Object) {
		open  = chars.open  || undefined;
		close = chars.close || undefined;
		esc   = chars.esc   || undefined;
	}
	
	if (Number.isInteger(lengths))
		min = lengths;
	
	else if (lengths instanceof Array) {
		min = lengths[0] || undefined;
		max = lengths[1] || undefined;
	}
	
	else if (lengths instanceof Object) {
		min = lengths.min || undefined;
		max = lengths.max || undefined;
	}
	
	// Regex escape the chars or set default
	esc   = typeof esc === 'string'   ? esc.replace(/[-\/\\^$*+?.()|[\]{}]/g,   '\\$&') : '\\\\';
	open  = typeof open === 'string'  ? open.replace(/[-\/\\^$*+?.()|[\]{}]/g,  '\\$&') : '\\1';
	close = typeof close === 'string' ? close.replace(/[-\/\\^$*+?.()|[\]{}]/g, '\\$&') : '\'"';
	
	// const evenEsc = '(?:[' + esc + ']{2})*(?![' + esc + '])';
	const oddEsc  = '(?:[' + esc + ']{2})*[' + esc + '](?![' + esc + '])';

	const approve  = '(?=' + oddEsc + ')';
	// Approve with '\' as the escape char
	//	(?=			# Positive lookahead
	//		(?:[\\]{2})*	# Match Pairs of escape char
	//		[\\]		# Match single escape
	//		(?![\\])	# Ensure no following escape char
	//	)
	
	const negate  = '(?!' + oddEsc + ')';
	// Negate with '\' as the escape char
	//	(?!			# Negative lookahead
	//		(?:[\\]{2})*	# Match Pairs of escape char
	//		[\\]		# Match single escape
	//		(?![\\])	# Ensure no following escape char
	//	)
	
	if (typeof internals !== 'string') {
		if (open === '\\1') {
			internals = (
				'(?:' +
					'(?:\\1' + approve + ')|' + 
					'(?:(?!\\1).)' +
				')'
			);
		} else {
			internals = (
				'(?:' +
					'(?:[' + close + ']' + approve + ')|' + 
					'(?:[' + open + ']' + approve + ')|' + 
					'(?:[^' + open + close + '])' +
				')'
			);
		}
	}
	
	// Open is a back reference
	//	(?:
	//		(?:
	//			\1	# Match any close quote
	//			approve	# Not followed by an escape
	//		)|
	//		(?:(?!\1).)	# Match any char that is no the close quote
	//	)
	
	// Open is not a back reference
	//	(?:
	//		(?:[“](?!(?=[\\]{2})*[\\](?![\\])))|	# Allow '“' as long as it's escaped with '\'
	//		(?:[”](?!(?=[\\]{2})*[\\](?![\\])))|	# Allow '”' as long as it's escaped with '\'
	//		(?:[^“”])				# Match anything that is not '“' or '”'
	//	)
	
	const length = (
		'{' +
			(min || '0') +
		',' +
			(max || '') +
		'}'
	);
	// Min = undefined, Max = undefined
	//	{0,}	# Length >= 0

	// Min = 5, Max = undefined
	//	{5,}	# Length >= 5
	
	// Min = undefined, Max = 10
	//	{0,10}	# 0 <= Length <= 10
	
	// Min = 5, Max = 10
	//	{5,10}	# 5 <= Length <= 10
	
	const regex = new RegExp(
		// Match the closing quote
		'([' + close + '])' + negate +
		// Match the string inside the quotes
		'(' + internals + length + ')' +
		// Match the opening quote
		'(' + (open === '\\1' ? '\\1' : ('[' + open + ']')) + ')' + negate,
		'g'
	);
	
	while(match = regex.exec(reversed)) {
		matches.push(
			match[2]
			.split('')
			.reverse()
			.join('')
			.replace(new RegExp(esc + match[1].replace(/[-\/\\^$*+?.()|[\]{}]/g, '\\$&'), 'g'), match[1])
			.replace(new RegExp(esc + match[3].replace(/[-\/\\^$*+?.()|[\]{}]/g, '\\$&'), 'g'), match[3])
		);
	}
	
	return matches;
}

matchQuotes.smaller.js

function matchQuotes(string, chars, lens, internals) {
	const reversed = string.split('').reverse().join('');
	const matches = [];
	let match;
	let open, close, esc;
	let min, max;
	
	open  = typeof chars === 'string' ? null  : (typeof chars == 'object' ? chars.open  || chars[0] : null);
	close = typeof chars === 'string' ? chars : (typeof chars == 'object' ? chars.close || chars[1] : null);
	esc   = typeof chars === 'string' ? null  : (typeof chars == 'object' ? chars.esc   || chars[2] : null);
	
	min = Number.isInteger(lens) ? lens : (typeof lens == 'object' ? (lens.min || lens[0] || '0') : '0');
	max = Number.isInteger(lens) ? ''   : (typeof lens == 'object' ? (lens.max || lens[1] || '' ) : '');
	
	esc   = typeof esc === 'string'   ? esc.replace(/[-\/\\^$*+?.()|[\]{}]/g,   '\\$&') : '\\\\';
	open  = typeof open === 'string'  ? open.replace(/[-\/\\^$*+?.()|[\]{}]/g,  '\\$&') : '\\1';
	close = typeof close === 'string' ? close.replace(/[-\/\\^$*+?.()|[\]{}]/g, '\\$&') : '\'"';

	const oddEsc  = '(?:[' + esc + ']{2})*[' + esc + '](?![' + esc + '])';
	const approve = '(?=' + oddEsc + ')';
	
	internals = typeof internals === 'string' ? internals : (
		open === '\\1' ? '(?:(?:\\1' + approve + ')|(?:(?!\\1).))' : (
			'(?:(?:['+close+']'+approve+')|(?:['+open+']'+approve+')|(?:[^'+open+close+']))'
		)
	);
	
	open = open === '\\1' ? '\\1' : '[' + open + ']';
	const regex = new RegExp(
		'([' + close + '])(?!' + oddEsc + ')' +
		'(' + internals + '{' + min + ',' + max + '}' + ')' +
		'(' + open + ')(?!' + oddEsc + ')',
		'g'
	);
	
	while(match = regex.exec(reversed))
		matches.push(
			match[2].split('').reverse().join('')
			.replace(new RegExp(esc + match[1].replace(/[-\/\\^$*+?.()|[\]{}]/g, '\\$&'), 'g'), match[1])
			.replace(new RegExp(esc + match[3].replace(/[-\/\\^$*+?.()|[\]{}]/g, '\\$&'), 'g'), match[3])
		);
	
	return matches;
}

matchQuotes.smallest.js

function matchQuotes(a,b,c,d){const e=a.split('').reverse().join(''),f=[];let g,h,i,j,k,l;h='string'==typeof b?null:'object'==typeof b?b.open||b[0]:null,i='string'==typeof b?b:'object'==typeof b?b.close||b[1]:null,j='string'==typeof b?null:'object'==typeof b?b.esc||b[2]:null,k=Number.isInteger(c)?c:'object'==typeof c?c.min||c[0]||'0':'0',l=Number.isInteger(c)?'':'object'==typeof c?c.max||c[1]||'':'',j='string'==typeof j?j.replace(/[-\/\\^$*+?.()|[\]{}]/g,'\\$&'):'\\\\',h='string'==typeof h?h.replace(/[-\/\\^$*+?.()|[\]{}]/g,'\\$&'):'\\1',i='string'==typeof i?i.replace(/[-\/\\^$*+?.()|[\]{}]/g,'\\$&'):'\'"';const m='(?:['+j+']{2})*['+j+'](?!['+j+'])',n='(?='+m+')';d='string'==typeof d?d:'\\1'===h?'(?:(?:\\1'+n+')|(?:(?!\\1).))':'(?:(?:['+i+']'+n+')|(?:['+h+']'+n+')|(?:[^'+h+i+']))',h='\\1'===h?'\\1':'['+h+']';for(const o=new RegExp('(['+i+'])(?!'+m+')('+d+'{'+k+','+l+'})('+h+')(?!'+m+')','g');g=o.exec(e);)f.push(g[2].split('').reverse().join('').replace(new RegExp(j+g[1].replace(/[-\/\\^$*+?.()|[\]{}]/g,'\\$&'),'g'),g[1]).replace(new RegExp(j+g[3].replace(/[-\/\\^$*+?.()|[\]{}]/g,'\\$&'),'g'),g[3]));return f}