single-file-core 1.0.0

package/modules/html-minifier.js

@@ -0,0 +1,262 @@
+/*
+ * Copyright 2010-2020 Gildas Lormeau
+ * contact : gildas.lormeau <at> gmail.com
+ * 
+ * This file is part of SingleFile.
+ *
+ *   The code in this file is free software: you can redistribute it and/or 
+ *   modify it under the terms of the GNU Affero General Public License 
+ *   (GNU AGPL) as published by the Free Software Foundation, either version 3
+ *   of the License, or (at your option) any later version.
+ * 
+ *   The code in this file is distributed in the hope that it will be useful, 
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of 
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero 
+ *   General Public License for more details.
+ *
+ *   As additional permission under GNU AGPL version 3 section 7, you may 
+ *   distribute UNMODIFIED VERSIONS OF THIS file without the copy of the GNU 
+ *   AGPL normally required by section 4, provided you include this license 
+ *   notice and a URL through which recipients can access the Corresponding 
+ *   Source.
+ */
+
+// Derived from the work of Kirill Maltsev - https://github.com/posthtml/htmlnano
+
+// Source: https://github.com/kangax/html-minifier/issues/63
+const booleanAttributes = [
+	"allowfullscreen",
+	"async",
+	"autofocus",
+	"autoplay",
+	"checked",
+	"compact",
+	"controls",
+	"declare",
+	"default",
+	"defaultchecked",
+	"defaultmuted",
+	"defaultselected",
+	"defer",
+	"disabled",
+	"enabled",
+	"formnovalidate",
+	"hidden",
+	"indeterminate",
+	"inert",
+	"ismap",
+	"itemscope",
+	"loop",
+	"multiple",
+	"muted",
+	"nohref",
+	"noresize",
+	"noshade",
+	"novalidate",
+	"nowrap",
+	"open",
+	"pauseonexit",
+	"readonly",
+	"required",
+	"reversed",
+	"scoped",
+	"seamless",
+	"selected",
+	"sortable",
+	"truespeed",
+	"typemustmatch",
+	"visible"
+];
+
+const noWhitespaceCollapseElements = ["script", "style", "pre", "textarea"];
+
+// Source: https://www.w3.org/TR/html4/sgml/dtd.html#events (Generic Attributes)
+const safeToRemoveAttrs = [
+	"id",
+	"class",
+	"style",
+	"lang",
+	"dir",
+	"onclick",
+	"ondblclick",
+	"onmousedown",
+	"onmouseup",
+	"onmouseover",
+	"onmousemove",
+	"onmouseout",
+	"onkeypress",
+	"onkeydown",
+	"onkeyup"
+];
+
+const redundantAttributes = {
+	"form": {
+		"method": "get"
+	},
+	"script": {
+		"language": "javascript",
+		"type": "text/javascript",
+		// Remove attribute if the function returns false
+		"charset": node => {
+			// The charset attribute only really makes sense on “external” SCRIPT elements:
+			// http://perfectionkills.com/optimizing-html/#8_script_charset
+			return !node.getAttribute("src");
+		}
+	},
+	"style": {
+		"media": "all",
+		"type": "text/css"
+	},
+	"link": {
+		"media": "all"
+	}
+};
+
+const REGEXP_WHITESPACE = /[ \t\f\r]+/g;
+const REGEXP_NEWLINE = /[\n]+/g;
+const REGEXP_ENDS_WHITESPACE = /^\s+$/;
+const NodeFilter_SHOW_ALL = 4294967295;
+const Node_ELEMENT_NODE = 1;
+const Node_TEXT_NODE = 3;
+const Node_COMMENT_NODE = 8;
+
+const modules = [
+	collapseBooleanAttributes,
+	mergeTextNodes,
+	collapseWhitespace,
+	removeComments,
+	removeEmptyAttributes,
+	removeRedundantAttributes,
+	compressJSONLD,
+	node => mergeElements(node, "style", (node, previousSibling) => node.parentElement && node.parentElement.tagName == "HEAD" && node.media == previousSibling.media && node.title == previousSibling.title)
+];
+
+export {
+	process
+};
+
+function process(doc, options) {
+	removeEmptyInlineElements(doc);
+	const nodesWalker = doc.createTreeWalker(doc.documentElement, NodeFilter_SHOW_ALL, null, false);
+	let node = nodesWalker.nextNode();
+	while (node) {
+		const deletedNode = modules.find(module => module(node, options));
+		const previousNode = node;
+		node = nodesWalker.nextNode();
+		if (deletedNode) {
+			previousNode.remove();
+		}
+	}
+}
+
+function collapseBooleanAttributes(node) {
+	if (node.nodeType == Node_ELEMENT_NODE) {
+		Array.from(node.attributes).forEach(attribute => {
+			if (booleanAttributes.includes(attribute.name)) {
+				node.setAttribute(attribute.name, "");
+			}
+		});
+	}
+}
+
+function mergeTextNodes(node) {
+	if (node.nodeType == Node_TEXT_NODE) {
+		if (node.previousSibling && node.previousSibling.nodeType == Node_TEXT_NODE) {
+			node.textContent = node.previousSibling.textContent + node.textContent;
+			node.previousSibling.remove();
+		}
+	}
+}
+
+function mergeElements(node, tagName, acceptMerge) {
+	if (node.nodeType == Node_ELEMENT_NODE && node.tagName.toLowerCase() == tagName.toLowerCase()) {
+		let previousSibling = node.previousSibling;
+		const previousSiblings = [];
+		while (previousSibling && previousSibling.nodeType == Node_TEXT_NODE && !previousSibling.textContent.trim()) {
+			previousSiblings.push(previousSibling);
+			previousSibling = previousSibling.previousSibling;
+		}
+		if (previousSibling && previousSibling.nodeType == Node_ELEMENT_NODE && previousSibling.tagName == node.tagName && acceptMerge(node, previousSibling)) {
+			node.textContent = previousSibling.textContent + node.textContent;
+			previousSiblings.forEach(node => node.remove());
+			previousSibling.remove();
+		}
+	}
+}
+
+function collapseWhitespace(node, options) {
+	if (node.nodeType == Node_TEXT_NODE) {
+		let element = node.parentElement;
+		const spacePreserved = element.getAttribute(options.PRESERVED_SPACE_ELEMENT_ATTRIBUTE_NAME) == "";
+		if (!spacePreserved) {
+			const textContent = node.textContent;
+			let noWhitespace = noWhitespaceCollapse(element);
+			while (noWhitespace) {
+				element = element.parentElement;
+				noWhitespace = element && noWhitespaceCollapse(element);
+			}
+			if ((!element || noWhitespace) && textContent.length > 1) {
+				node.textContent = textContent.replace(REGEXP_WHITESPACE, getWhiteSpace(node)).replace(REGEXP_NEWLINE, "\n");
+			}
+		}
+	}
+}
+
+function getWhiteSpace(node) {
+	return node.parentElement && node.parentElement.tagName == "HEAD" ? "\n" : " ";
+}
+
+function noWhitespaceCollapse(element) {
+	return element && !noWhitespaceCollapseElements.includes(element.tagName.toLowerCase());
+}
+
+function removeComments(node) {
+	if (node.nodeType == Node_COMMENT_NODE && node.parentElement.tagName != "HTML") {
+		return !node.textContent.toLowerCase().trim().startsWith("[if");
+	}
+}
+
+function removeEmptyAttributes(node) {
+	if (node.nodeType == Node_ELEMENT_NODE) {
+		Array.from(node.attributes).forEach(attribute => {
+			if (safeToRemoveAttrs.includes(attribute.name.toLowerCase())) {
+				const attributeValue = node.getAttribute(attribute.name);
+				if (attributeValue == "" || (attributeValue || "").match(REGEXP_ENDS_WHITESPACE)) {
+					node.removeAttribute(attribute.name);
+				}
+			}
+		});
+	}
+}
+
+function removeRedundantAttributes(node) {
+	if (node.nodeType == Node_ELEMENT_NODE) {
+		const tagRedundantAttributes = redundantAttributes[node.tagName.toLowerCase()];
+		if (tagRedundantAttributes) {
+			Object.keys(tagRedundantAttributes).forEach(redundantAttributeName => {
+				const tagRedundantAttributeValue = tagRedundantAttributes[redundantAttributeName];
+				if (typeof tagRedundantAttributeValue == "function" ? tagRedundantAttributeValue(node) : node.getAttribute(redundantAttributeName) == tagRedundantAttributeValue) {
+					node.removeAttribute(redundantAttributeName);
+				}
+			});
+		}
+	}
+}
+
+function compressJSONLD(node) {
+	if (node.nodeType == Node_ELEMENT_NODE && node.tagName == "SCRIPT" && node.type == "application/ld+json" && node.textContent.trim()) {
+		try {
+			node.textContent = JSON.stringify(JSON.parse(node.textContent));
+		} catch (error) {
+			// ignored
+		}
+	}
+}
+
+function removeEmptyInlineElements(doc) {
+	doc.querySelectorAll("style, script:not([src])").forEach(element => {
+		if (!element.textContent.trim()) {
+			element.remove();
+		}
+	});
+}

package/modules/html-serializer.js

@@ -0,0 +1,180 @@
+/*
+ * Copyright 2010-2020 Gildas Lormeau
+ * contact : gildas.lormeau <at> gmail.com
+ * 
+ * This file is part of SingleFile.
+ *
+ *   The code in this file is free software: you can redistribute it and/or 
+ *   modify it under the terms of the GNU Affero General Public License 
+ *   (GNU AGPL) as published by the Free Software Foundation, either version 3
+ *   of the License, or (at your option) any later version.
+ * 
+ *   The code in this file is distributed in the hope that it will be useful, 
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of 
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero 
+ *   General Public License for more details.
+ *
+ *   As additional permission under GNU AGPL version 3 section 7, you may 
+ *   distribute UNMODIFIED VERSIONS OF THIS file without the copy of the GNU 
+ *   AGPL normally required by section 4, provided you include this license 
+ *   notice and a URL through which recipients can access the Corresponding 
+ *   Source.
+ */
+
+const SELF_CLOSED_TAG_NAMES = ["area", "base", "br", "col", "command", "embed", "hr", "img", "input", "keygen", "link", "meta", "param", "source", "track", "wbr"];
+
+const Node_ELEMENT_NODE = 1;
+const Node_TEXT_NODE = 3;
+const Node_COMMENT_NODE = 8;
+
+// see https://www.w3.org/TR/html5/syntax.html#optional-tags
+const OMITTED_START_TAGS = [
+	{ tagName: "head", accept: element => !element.childNodes.length || element.childNodes[0].nodeType == Node_ELEMENT_NODE },
+	{ tagName: "body", accept: element => !element.childNodes.length }
+];
+const OMITTED_END_TAGS = [
+	{ tagName: "html", accept: next => !next || next.nodeType != Node_COMMENT_NODE },
+	{ tagName: "head", accept: next => !next || (next.nodeType != Node_COMMENT_NODE && (next.nodeType != Node_TEXT_NODE || !startsWithSpaceChar(next.textContent))) },
+	{ tagName: "body", accept: next => !next || next.nodeType != Node_COMMENT_NODE },
+	{ tagName: "li", accept: (next, element) => (!next && element.parentElement && (element.parentElement.tagName == "UL" || element.parentElement.tagName == "OL")) || (next && ["LI"].includes(next.tagName)) },
+	{ tagName: "dt", accept: next => !next || ["DT", "DD"].includes(next.tagName) },
+	{ tagName: "p", accept: next => next && ["ADDRESS", "ARTICLE", "ASIDE", "BLOCKQUOTE", "DETAILS", "DIV", "DL", "FIELDSET", "FIGCAPTION", "FIGURE", "FOOTER", "FORM", "H1", "H2", "H3", "H4", "H5", "H6", "HEADER", "HR", "MAIN", "NAV", "OL", "P", "PRE", "SECTION", "TABLE", "UL"].includes(next.tagName) },
+	{ tagName: "dd", accept: next => !next || ["DT", "DD"].includes(next.tagName) },
+	{ tagName: "rt", accept: next => !next || ["RT", "RP"].includes(next.tagName) },
+	{ tagName: "rp", accept: next => !next || ["RT", "RP"].includes(next.tagName) },
+	{ tagName: "optgroup", accept: next => !next || ["OPTGROUP"].includes(next.tagName) },
+	{ tagName: "option", accept: next => !next || ["OPTION", "OPTGROUP"].includes(next.tagName) },
+	{ tagName: "colgroup", accept: next => !next || (next.nodeType != Node_COMMENT_NODE && (next.nodeType != Node_TEXT_NODE || !startsWithSpaceChar(next.textContent))) },
+	{ tagName: "caption", accept: next => !next || (next.nodeType != Node_COMMENT_NODE && (next.nodeType != Node_TEXT_NODE || !startsWithSpaceChar(next.textContent))) },
+	{ tagName: "thead", accept: next => !next || ["TBODY", "TFOOT"].includes(next.tagName) },
+	{ tagName: "tbody", accept: next => !next || ["TBODY", "TFOOT"].includes(next.tagName) },
+	{ tagName: "tfoot", accept: next => !next },
+	{ tagName: "tr", accept: next => !next || ["TR"].includes(next.tagName) },
+	{ tagName: "td", accept: next => !next || ["TD", "TH"].includes(next.tagName) },
+	{ tagName: "th", accept: next => !next || ["TD", "TH"].includes(next.tagName) }
+];
+const TEXT_NODE_TAGS = ["style", "script", "xmp", "iframe", "noembed", "noframes", "plaintext", "noscript"];
+
+export {
+	process
+};
+
+function process(doc, compressHTML) {
+	const docType = doc.doctype;
+	let docTypeString = "";
+	if (docType) {
+		docTypeString = "<!DOCTYPE " + docType.nodeName;
+		if (docType.publicId) {
+			docTypeString += " PUBLIC \"" + docType.publicId + "\"";
+			if (docType.systemId)
+				docTypeString += " \"" + docType.systemId + "\"";
+		} else if (docType.systemId)
+			docTypeString += " SYSTEM \"" + docType.systemId + "\"";
+		if (docType.internalSubset)
+			docTypeString += " [" + docType.internalSubset + "]";
+		docTypeString += "> ";
+	}
+	return docTypeString + serialize(doc.documentElement, compressHTML);
+}
+
+function serialize(node, compressHTML, isSVG) {
+	if (node.nodeType == Node_TEXT_NODE) {
+		return serializeTextNode(node);
+	} else if (node.nodeType == Node_COMMENT_NODE) {
+		return serializeCommentNode(node);
+	} else if (node.nodeType == Node_ELEMENT_NODE) {
+		return serializeElement(node, compressHTML, isSVG);
+	}
+}
+
+function serializeTextNode(textNode) {
+	const parentNode = textNode.parentNode;
+	let parentTagName;
+	if (parentNode && parentNode.nodeType == Node_ELEMENT_NODE) {
+		parentTagName = parentNode.tagName.toLowerCase();
+	}
+	if (!parentTagName || TEXT_NODE_TAGS.includes(parentTagName)) {
+		if (parentTagName == "script") {
+			return textNode.textContent.replace(/<\//gi, "<\\/").replace(/\/>/gi, "\\/>");
+		}
+		return textNode.textContent;
+	} else {
+		return textNode.textContent.replace(/&/g, "&amp;").replace(/\u00a0/g, "&nbsp;").replace(/</g, "&lt;").replace(/>/g, "&gt;");
+	}
+}
+
+function serializeCommentNode(commentNode) {
+	return "<!--" + commentNode.textContent + "-->";
+}
+
+function serializeElement(element, compressHTML, isSVG) {
+	const tagName = element.tagName.toLowerCase();
+	const omittedStartTag = compressHTML && OMITTED_START_TAGS.find(omittedStartTag => tagName == omittedStartTag.tagName && omittedStartTag.accept(element));
+	let content = "";
+	if (!omittedStartTag || element.attributes.length) {
+		content = "<" + tagName;
+		Array.from(element.attributes).forEach(attribute => content += serializeAttribute(attribute, element, compressHTML));
+		content += ">";
+	}
+	if (element.tagName == "TEMPLATE" && !element.childNodes.length) {
+		content += element.innerHTML;
+	} else {
+		Array.from(element.childNodes).forEach(childNode => content += serialize(childNode, compressHTML, isSVG || tagName == "svg"));
+	}
+	const omittedEndTag = compressHTML && OMITTED_END_TAGS.find(omittedEndTag => tagName == omittedEndTag.tagName && omittedEndTag.accept(element.nextSibling, element));
+	if (isSVG || (!omittedEndTag && !SELF_CLOSED_TAG_NAMES.includes(tagName))) {
+		content += "</" + tagName + ">";
+	}
+	return content;
+}
+
+function serializeAttribute(attribute, element, compressHTML) {
+	const name = attribute.name;
+	let content = "";
+	if (!name.match(/["'>/=]/)) {
+		let value = attribute.value;
+		if (compressHTML && name == "class") {
+			value = Array.from(element.classList).map(className => className.trim()).join(" ");
+		}
+		let simpleQuotesValue;
+		value = value.replace(/&/g, "&amp;").replace(/\u00a0/g, "&nbsp;");
+		if (value.includes("\"")) {
+			if (value.includes("'") || !compressHTML) {
+				value = value.replace(/"/g, "&quot;");
+			} else {
+				simpleQuotesValue = true;
+			}
+		}
+		const invalidUnquotedValue = !compressHTML || !value.match(/^[^ \t\n\f\r'"`=<>]+$/);
+		content += " ";
+		if (!attribute.namespace) {
+			content += name;
+		} else if (attribute.namespaceURI == "http://www.w3.org/XML/1998/namespace") {
+			content += "xml:" + name;
+		} else if (attribute.namespaceURI == "http://www.w3.org/2000/xmlns/") {
+			if (name !== "xmlns") {
+				content += "xmlns:";
+			}
+			content += name;
+		} else if (attribute.namespaceURI == "http://www.w3.org/1999/xlink") {
+			content += "xlink:" + name;
+		} else {
+			content += name;
+		}
+		if (value != "") {
+			content += "=";
+			if (invalidUnquotedValue) {
+				content += simpleQuotesValue ? "'" : "\"";
+			}
+			content += value;
+			if (invalidUnquotedValue) {
+				content += simpleQuotesValue ? "'" : "\"";
+			}
+		}
+	}
+	return content;
+}
+
+function startsWithSpaceChar(textContent) {
+	return Boolean(textContent.match(/^[ \t\n\f\r]/));
+}

package/modules/index.js

@@ -0,0 +1,42 @@
+/*
+ * Copyright 2010-2020 Gildas Lormeau
+ * contact : gildas.lormeau <at> gmail.com
+ * 
+ * This file is part of SingleFile.
+ *
+ *   The code in this file is free software: you can redistribute it and/or 
+ *   modify it under the terms of the GNU Affero General Public License 
+ *   (GNU AGPL) as published by the Free Software Foundation, either version 3
+ *   of the License, or (at your option) any later version.
+ * 
+ *   The code in this file is distributed in the hope that it will be useful, 
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of 
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero 
+ *   General Public License for more details.
+ *
+ *   As additional permission under GNU AGPL version 3 section 7, you may 
+ *   distribute UNMODIFIED VERSIONS OF THIS file without the copy of the GNU 
+ *   AGPL normally required by section 4, provided you include this license 
+ *   notice and a URL through which recipients can access the Corresponding 
+ *   Source.
+ */
+
+import * as fontsAltMinifier from "./css-fonts-alt-minifier.js";
+import * as fontsMinifier from "./css-fonts-minifier";
+import * as matchedRules from "./css-matched-rules.js";
+import * as mediasAltMinifier from "./css-medias-alt-minifier.js";
+import * as cssRulesMinifier from "./css-rules-minifier.js";
+import * as imagesAltMinifier from "./html-images-alt-minifier.js";
+import * as htmlMinifier from "./html-minifier.js";
+import * as serializer from "./html-serializer.js";
+
+export {
+	fontsAltMinifier,
+	fontsMinifier,
+	matchedRules,
+	mediasAltMinifier,
+	cssRulesMinifier,
+	imagesAltMinifier,
+	htmlMinifier,
+	serializer
+};

package/package.json

@@ -0,0 +1,18 @@
+{
+  "name": "single-file-core",
+  "version": "1.0.0",
+  "description": "SingleFile Core",
+  "author": "Gildas Lormeau",
+  "license": "AGPL-3.0-or-later",
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/gildas-lormeau/single-file-core.git"
+  },
+  "bugs": {
+    "url": "https://github.com/gildas-lormeau/single-file-core/issues"
+  },
+  "homepage": "https://github.com/gildas-lormeau/single-file-core#readme"
+}