SingleFile/lib/single-file/html-srcset-parser.js

/**
 * Srcset Parser
 *
 * By Alex Bell |  MIT License
 *
 * JS Parser for the string value that appears in markup <img srcset="here">
 *
 * @returns Array [{url: _, d: _, w: _, h:_}, ...]
 *
 * Based super duper closely on the reference algorithm at:
 * https://html.spec.whatwg.org/multipage/embedded-content.html#parse-a-srcset-attribute
 *
 * Most comments are copied in directly from the spec
 * (except for comments in parens).
 */

this.srcsetParser = this.srcsetParser || (() => {

	return {
		process
	};

	// 1. Let input be the value passed to this algorithm.
	function process(input) {

		// UTILITY FUNCTIONS

		// Manual is faster than RegEx
		// http://bjorn.tipling.com/state-and-regular-expressions-in-javascript
		// http://jsperf.com/whitespace-character/5
		function isSpace(c) {
			return (c === "\u0020" || // space
				c === "\u0009" || // horizontal tab
				c === "\u000A" || // new line
				c === "\u000C" || // form feed
				c === "\u000D");  // carriage return
		}

		function collectCharacters(regEx) {
			let chars;
			const match = regEx.exec(input.substring(pos));
			if (match) {
				chars = match[0];
				pos += chars.length;
				return chars;
			}
		}

		const inputLength = input.length;

		// (Don"t use \s, to avoid matching non-breaking space)
		/* eslint-disable no-control-regex */
		const regexLeadingSpaces = /^[ \t\n\r\u000c]+/;
		const regexLeadingCommasOrSpaces = /^[, \t\n\r\u000c]+/;
		const regexLeadingNotSpaces = /^[^ \t\n\r\u000c]+/;
		const regexTrailingCommas = /[,]+$/;
		const regexNonNegativeInteger = /^\d+$/;
		/* eslint-enable no-control-regex */

		// ( Positive or negative or unsigned integers or decimals, without or without exponents.
		// Must include at least one digit.
		// According to spec tests any decimal point must be followed by a digit.
		// No leading plus sign is allowed.)
		// https://html.spec.whatwg.org/multipage/infrastructure.html#valid-floating-point-number
		const regexFloatingPoint = /^-?(?:[0-9]+|[0-9]*\.[0-9]+)(?:[eE][+-]?[0-9]+)?$/;

		let url, descriptors, currentDescriptor, state, c,
			// 2. Let position be a pointer into input, initially pointing at the start
			//    of the string.
			pos = 0;
		// 3. Let candidates be an initially empty source set.
		const candidates = [];

		// 4. Splitting loop: Collect a sequence of characters that are space
		//    characters or U+002C COMMA characters. If any U+002C COMMA characters
		//    were collected, that is a parse error.		
		while (true) { // eslint-disable-line no-constant-condition
			collectCharacters(regexLeadingCommasOrSpaces);

			// 5. If position is past the end of input, return candidates and abort these steps.
			if (pos >= inputLength) {
				return candidates; // (we"re done, this is the sole return path)
			}

			// 6. Collect a sequence of characters that are not space characters,
			//    and let that be url.
			url = collectCharacters(regexLeadingNotSpaces);

			// 7. Let descriptors be a new empty list.
			descriptors = [];

			// 8. If url ends with a U+002C COMMA character (,), follow these substeps:
			//		(1). Remove all trailing U+002C COMMA characters from url. If this removed
			//         more than one character, that is a parse error.
			if (url.slice(-1) === ",") {
				url = url.replace(regexTrailingCommas, "");
				// (Jump ahead to step 9 to skip tokenization and just push the candidate).
				parseDescriptors();

				//	Otherwise, follow these substeps:
			} else {
				tokenize();
			} // (close else of step 8)

			// 16. Return to the step labeled splitting loop.
		} // (Close of big while loop.)

		/**
		 * Tokenizes descriptor properties prior to parsing
		 * Returns undefined.
		 */
		function tokenize() {

			// 8.1. Descriptor tokeniser: Skip whitespace
			collectCharacters(regexLeadingSpaces);

			// 8.2. Let current descriptor be the empty string.
			currentDescriptor = "";

			// 8.3. Let state be in descriptor.
			state = "in descriptor";

			while (true) { // eslint-disable-line no-constant-condition

				// 8.4. Let c be the character at position.
				c = input.charAt(pos);

				//  Do the following depending on the value of state.
				//  For the purpose of this step, "EOF" is a special character representing
				//  that position is past the end of input.

				// In descriptor
				if (state === "in descriptor") {
					// Do the following, depending on the value of c:

					// Space character
					// If current descriptor is not empty, append current descriptor to
					// descriptors and let current descriptor be the empty string.
					// Set state to after descriptor.
					if (isSpace(c)) {
						if (currentDescriptor) {
							descriptors.push(currentDescriptor);
							currentDescriptor = "";
							state = "after descriptor";
						}

						// U+002C COMMA (,)
						// Advance position to the next character in input. If current descriptor
						// is not empty, append current descriptor to descriptors. Jump to the step
						// labeled descriptor parser.
					} else if (c === ",") {
						pos += 1;
						if (currentDescriptor) {
							descriptors.push(currentDescriptor);
						}
						parseDescriptors();
						return;

						// U+0028 LEFT PARENTHESIS (()
						// Append c to current descriptor. Set state to in parens.
					} else if (c === "\u0028") {
						currentDescriptor = currentDescriptor + c;
						state = "in parens";

						// EOF
						// If current descriptor is not empty, append current descriptor to
						// descriptors. Jump to the step labeled descriptor parser.
					} else if (c === "") {
						if (currentDescriptor) {
							descriptors.push(currentDescriptor);
						}
						parseDescriptors();
						return;

						// Anything else
						// Append c to current descriptor.
					} else {
						currentDescriptor = currentDescriptor + c;
					}
					// (end "in descriptor"

					// In parens
				} else if (state === "in parens") {

					// U+0029 RIGHT PARENTHESIS ())
					// Append c to current descriptor. Set state to in descriptor.
					if (c === ")") {
						currentDescriptor = currentDescriptor + c;
						state = "in descriptor";

						// EOF
						// Append current descriptor to descriptors. Jump to the step labeled
						// descriptor parser.
					} else if (c === "") {
						descriptors.push(currentDescriptor);
						parseDescriptors();
						return;

						// Anything else
						// Append c to current descriptor.
					} else {
						currentDescriptor = currentDescriptor + c;
					}

					// After descriptor
				} else if (state === "after descriptor") {

					// Do the following, depending on the value of c:
					// Space character: Stay in this state.
					if (isSpace(c)) {

						// EOF: Jump to the step labeled descriptor parser.
					} else if (c === "") {
						parseDescriptors();
						return;

						// Anything else
						// Set state to in descriptor. Set position to the previous character in input.
					} else {
						state = "in descriptor";
						pos -= 1;

					}
				}

				// Advance position to the next character in input.
				pos += 1;

				// Repeat this step.
			} // (close while true loop)
		}

		/**
		 * Adds descriptor properties to a candidate, pushes to the candidates array
		 * @return undefined
		 */
		// Declared outside of the while loop so that it"s only created once.
		function parseDescriptors() {

			// 9. Descriptor parser: Let error be no.
			let pError = false,

				// 10. Let width be absent.
				// 11. Let density be absent.
				// 12. Let future-compat-h be absent. (We"re implementing it now as h)
				w, d, h, i,
				desc, lastChar, value, intVal, floatVal;
			const candidate = {};

			// 13. For each descriptor in descriptors, run the appropriate set of steps
			// from the following list:
			for (i = 0; i < descriptors.length; i++) {
				desc = descriptors[i];

				lastChar = desc[desc.length - 1];
				value = desc.substring(0, desc.length - 1);
				intVal = parseInt(value, 10);
				floatVal = parseFloat(value);

				// If the descriptor consists of a valid non-negative integer followed by
				// a U+0077 LATIN SMALL LETTER W character
				if (regexNonNegativeInteger.test(value) && (lastChar === "w")) {

					// If width and density are not both absent, then let error be yes.
					if (w || d) { pError = true; }

					// Apply the rules for parsing non-negative integers to the descriptor.
					// If the result is zero, let error be yes.
					// Otherwise, let width be the result.
					if (intVal === 0) { pError = true; } else { w = intVal; }

					// If the descriptor consists of a valid floating-point number followed by
					// a U+0078 LATIN SMALL LETTER X character
				} else if (regexFloatingPoint.test(value) && (lastChar === "x")) {

					// If width, density and future-compat-h are not all absent, then let error
					// be yes.
					if (w || d || h) { pError = true; }

					// Apply the rules for parsing floating-point number values to the descriptor.
					// If the result is less than zero, let error be yes. Otherwise, let density
					// be the result.
					if (floatVal < 0) { pError = true; } else { d = floatVal; }

					// If the descriptor consists of a valid non-negative integer followed by
					// a U+0068 LATIN SMALL LETTER H character
				} else if (regexNonNegativeInteger.test(value) && (lastChar === "h")) {

					// If height and density are not both absent, then let error be yes.
					if (h || d) { pError = true; }

					// Apply the rules for parsing non-negative integers to the descriptor.
					// If the result is zero, let error be yes. Otherwise, let future-compat-h
					// be the result.
					if (intVal === 0) { pError = true; } else { h = intVal; }

					// Anything else, Let error be yes.
				} else { pError = true; }
			} // (close step 13 for loop)

			// 15. If error is still no, then append a new image source to candidates whose
			// URL is url, associated with a width width if not absent and a pixel
			// density density if not absent. Otherwise, there is a parse error.
			if (!pError) {
				candidate.url = url;
				if (w) { candidate.w = w; }
				if (d) { candidate.d = d; }
				if (h) { candidate.h = h; }
				candidates.push(candidate);
			} else if (console && console.log) {  // eslint-disable-line no-console
				console.log("Invalid srcset descriptor found in \"" + input + "\" at \"" + desc + "\"."); // eslint-disable-line no-console
			}
		} // (close parseDescriptors fn)

	}

})();