diff options
author | Florian Dold <florian.dold@gmail.com> | 2017-05-03 15:35:00 +0200 |
---|---|---|
committer | Florian Dold <florian.dold@gmail.com> | 2017-05-03 15:35:00 +0200 |
commit | de98e0b232509d5f40c135d540a70e415272ff85 (patch) | |
tree | a79222a5b58484ab3b80d18efcaaa7ccc4769b33 /node_modules/html-minifier/src | |
parent | e0c9d480a73fa629c1e4a47d3e721f1d2d345406 (diff) |
node_modules
Diffstat (limited to 'node_modules/html-minifier/src')
-rw-r--r-- | node_modules/html-minifier/src/htmlminifier.js | 1308 | ||||
-rw-r--r-- | node_modules/html-minifier/src/htmlparser.js | 528 | ||||
-rw-r--r-- | node_modules/html-minifier/src/tokenchain.js | 71 | ||||
-rw-r--r-- | node_modules/html-minifier/src/utils.js | 18 |
4 files changed, 1925 insertions, 0 deletions
diff --git a/node_modules/html-minifier/src/htmlminifier.js b/node_modules/html-minifier/src/htmlminifier.js new file mode 100644 index 000000000..f7917cd4a --- /dev/null +++ b/node_modules/html-minifier/src/htmlminifier.js @@ -0,0 +1,1308 @@ +'use strict'; + +var CleanCSS = require('clean-css'); +var decode = require('he').decode; +var HTMLParser = require('./htmlparser').HTMLParser; +var RelateUrl = require('relateurl'); +var TokenChain = require('./tokenchain'); +var UglifyJS = require('uglify-js'); +var utils = require('./utils'); + +var trimWhitespace = String.prototype.trim ? function(str) { + if (typeof str !== 'string') { + return str; + } + return str.trim(); +} : function(str) { + if (typeof str !== 'string') { + return str; + } + return str.replace(/^\s+/, '').replace(/\s+$/, ''); +}; + +function collapseWhitespaceAll(str) { + return str && str.replace(/\s+/g, function(spaces) { + return spaces === '\t' ? '\t' : spaces.replace(/(^|\xA0+)[^\xA0]+/g, '$1 '); + }); +} + +function collapseWhitespace(str, options, trimLeft, trimRight, collapseAll) { + var lineBreakBefore = '', lineBreakAfter = ''; + + if (options.preserveLineBreaks) { + str = str.replace(/^\s*?[\n\r]\s*/, function() { + lineBreakBefore = '\n'; + return ''; + }).replace(/\s*?[\n\r]\s*$/, function() { + lineBreakAfter = '\n'; + return ''; + }); + } + + if (trimLeft) { + str = str.replace(/^\s+/, function(spaces) { + var conservative = !lineBreakBefore && options.conservativeCollapse; + if (conservative && spaces === '\t') { + return '\t'; + } + return spaces.replace(/^[^\xA0]+/, '').replace(/(\xA0+)[^\xA0]+/g, '$1 ') || (conservative ? ' ' : ''); + }); + } + + if (trimRight) { + str = str.replace(/\s+$/, function(spaces) { + var conservative = !lineBreakAfter && options.conservativeCollapse; + if (conservative && spaces === '\t') { + return '\t'; + } + return spaces.replace(/[^\xA0]+(\xA0+)/g, ' $1').replace(/[^\xA0]+$/, '') || (conservative ? ' ' : ''); + }); + } + + if (collapseAll) { + // strip non space whitespace then compress spaces to one + str = collapseWhitespaceAll(str); + } + + return lineBreakBefore + str + lineBreakAfter; +} + +var createMapFromString = utils.createMapFromString; +// non-empty tags that will maintain whitespace around them +var inlineTags = createMapFromString('a,abbr,acronym,b,bdi,bdo,big,button,cite,code,del,dfn,em,font,i,ins,kbd,mark,math,nobr,q,rt,rp,s,samp,small,span,strike,strong,sub,sup,svg,time,tt,u,var'); +// non-empty tags that will maintain whitespace within them +var inlineTextTags = createMapFromString('a,abbr,acronym,b,big,del,em,font,i,ins,kbd,mark,nobr,s,samp,small,span,strike,strong,sub,sup,time,tt,u,var'); +// self-closing tags that will maintain whitespace around them +var selfClosingInlineTags = createMapFromString('comment,img,input,wbr'); + +function collapseWhitespaceSmart(str, prevTag, nextTag, options) { + var trimLeft = prevTag && !selfClosingInlineTags(prevTag); + if (trimLeft && !options.collapseInlineTagWhitespace) { + trimLeft = prevTag.charAt(0) === '/' ? !inlineTags(prevTag.slice(1)) : !inlineTextTags(prevTag); + } + var trimRight = nextTag && !selfClosingInlineTags(nextTag); + if (trimRight && !options.collapseInlineTagWhitespace) { + trimRight = nextTag.charAt(0) === '/' ? !inlineTextTags(nextTag.slice(1)) : !inlineTags(nextTag); + } + return collapseWhitespace(str, options, trimLeft, trimRight, prevTag && nextTag); +} + +function isConditionalComment(text) { + return /^\[if\s[^\]]+]|\[endif]$/.test(text); +} + +function isIgnoredComment(text, options) { + for (var i = 0, len = options.ignoreCustomComments.length; i < len; i++) { + if (options.ignoreCustomComments[i].test(text)) { + return true; + } + } + return false; +} + +function isEventAttribute(attrName, options) { + var patterns = options.customEventAttributes; + if (patterns) { + for (var i = patterns.length; i--;) { + if (patterns[i].test(attrName)) { + return true; + } + } + return false; + } + return /^on[a-z]{3,}$/.test(attrName); +} + +function canRemoveAttributeQuotes(value) { + // http://mathiasbynens.be/notes/unquoted-attribute-values + return /^[^ \t\n\f\r"'`=<>]+$/.test(value); +} + +function attributesInclude(attributes, attribute) { + for (var i = attributes.length; i--;) { + if (attributes[i].name.toLowerCase() === attribute) { + return true; + } + } + return false; +} + +function isAttributeRedundant(tag, attrName, attrValue, attrs) { + attrValue = attrValue ? trimWhitespace(attrValue.toLowerCase()) : ''; + + return ( + tag === 'script' && + attrName === 'language' && + attrValue === 'javascript' || + + tag === 'form' && + attrName === 'method' && + attrValue === 'get' || + + tag === 'input' && + attrName === 'type' && + attrValue === 'text' || + + tag === 'script' && + attrName === 'charset' && + !attributesInclude(attrs, 'src') || + + tag === 'a' && + attrName === 'name' && + attributesInclude(attrs, 'id') || + + tag === 'area' && + attrName === 'shape' && + attrValue === 'rect' + ); +} + +// https://mathiasbynens.be/demo/javascript-mime-type +// https://developer.mozilla.org/en/docs/Web/HTML/Element/script#attr-type +var executableScriptsMimetypes = utils.createMap([ + 'text/javascript', + 'text/ecmascript', + 'text/jscript', + 'application/javascript', + 'application/x-javascript', + 'application/ecmascript' +]); + +function isScriptTypeAttribute(attrValue) { + attrValue = trimWhitespace(attrValue.split(/;/, 2)[0]).toLowerCase(); + return attrValue === '' || executableScriptsMimetypes(attrValue); +} + +function isExecutableScript(tag, attrs) { + if (tag !== 'script') { + return false; + } + for (var i = 0, len = attrs.length; i < len; i++) { + var attrName = attrs[i].name.toLowerCase(); + if (attrName === 'type') { + return isScriptTypeAttribute(attrs[i].value); + } + } + return true; +} + +function isStyleLinkTypeAttribute(attrValue) { + attrValue = trimWhitespace(attrValue).toLowerCase(); + return attrValue === '' || attrValue === 'text/css'; +} + +function isStyleSheet(tag, attrs) { + if (tag !== 'style') { + return false; + } + for (var i = 0, len = attrs.length; i < len; i++) { + var attrName = attrs[i].name.toLowerCase(); + if (attrName === 'type') { + return isStyleLinkTypeAttribute(attrs[i].value); + } + } + return true; +} + +var isSimpleBoolean = createMapFromString('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'); +var isBooleanValue = createMapFromString('true,false'); + +function isBooleanAttribute(attrName, attrValue) { + return isSimpleBoolean(attrName) || attrName === 'draggable' && !isBooleanValue(attrValue); +} + +function isUriTypeAttribute(attrName, tag) { + return ( + /^(?:a|area|link|base)$/.test(tag) && attrName === 'href' || + tag === 'img' && /^(?:src|longdesc|usemap)$/.test(attrName) || + tag === 'object' && /^(?:classid|codebase|data|usemap)$/.test(attrName) || + tag === 'q' && attrName === 'cite' || + tag === 'blockquote' && attrName === 'cite' || + (tag === 'ins' || tag === 'del') && attrName === 'cite' || + tag === 'form' && attrName === 'action' || + tag === 'input' && (attrName === 'src' || attrName === 'usemap') || + tag === 'head' && attrName === 'profile' || + tag === 'script' && (attrName === 'src' || attrName === 'for') + ); +} + +function isNumberTypeAttribute(attrName, tag) { + return ( + /^(?:a|area|object|button)$/.test(tag) && attrName === 'tabindex' || + tag === 'input' && (attrName === 'maxlength' || attrName === 'tabindex') || + tag === 'select' && (attrName === 'size' || attrName === 'tabindex') || + tag === 'textarea' && /^(?:rows|cols|tabindex)$/.test(attrName) || + tag === 'colgroup' && attrName === 'span' || + tag === 'col' && attrName === 'span' || + (tag === 'th' || tag === 'td') && (attrName === 'rowspan' || attrName === 'colspan') + ); +} + +function isLinkType(tag, attrs, value) { + if (tag !== 'link') { + return false; + } + for (var i = 0, len = attrs.length; i < len; i++) { + if (attrs[i].name === 'rel' && attrs[i].value === value) { + return true; + } + } +} + +function isMediaQuery(tag, attrs, attrName) { + return attrName === 'media' && (isLinkType(tag, attrs, 'stylesheet') || isStyleSheet(tag, attrs)); +} + +var srcsetTags = createMapFromString('img,source'); + +function isSrcset(attrName, tag) { + return attrName === 'srcset' && srcsetTags(tag); +} + +function cleanAttributeValue(tag, attrName, attrValue, options, attrs) { + if (attrValue && isEventAttribute(attrName, options)) { + attrValue = trimWhitespace(attrValue).replace(/^javascript:\s*/i, ''); + return options.minifyJS(attrValue, true); + } + else if (attrName === 'class') { + attrValue = trimWhitespace(attrValue); + if (options.sortClassName) { + attrValue = options.sortClassName(attrValue); + } + else { + attrValue = collapseWhitespaceAll(attrValue); + } + return attrValue; + } + else if (isUriTypeAttribute(attrName, tag)) { + attrValue = trimWhitespace(attrValue); + return isLinkType(tag, attrs, 'canonical') ? attrValue : options.minifyURLs(attrValue); + } + else if (isNumberTypeAttribute(attrName, tag)) { + return trimWhitespace(attrValue); + } + else if (attrName === 'style') { + attrValue = trimWhitespace(attrValue); + if (attrValue) { + if (/;$/.test(attrValue) && !/&#?[0-9a-zA-Z]+;$/.test(attrValue)) { + attrValue = attrValue.replace(/\s*;$/, ''); + } + attrValue = unwrapInlineCSS(options.minifyCSS(wrapInlineCSS(attrValue))); + } + return attrValue; + } + else if (isSrcset(attrName, tag)) { + // https://html.spec.whatwg.org/multipage/embedded-content.html#attr-img-srcset + attrValue = trimWhitespace(attrValue).split(/\s+,\s*|\s*,\s+/).map(function(candidate) { + var url = candidate; + var descriptor = ''; + var match = candidate.match(/\s+([1-9][0-9]*w|[0-9]+(?:\.[0-9]+)?x)$/); + if (match) { + url = url.slice(0, -match[0].length); + var num = +match[1].slice(0, -1); + var suffix = match[1].slice(-1); + if (num !== 1 || suffix !== 'x') { + descriptor = ' ' + num + suffix; + } + } + return options.minifyURLs(url) + descriptor; + }).join(', '); + } + else if (isMetaViewport(tag, attrs) && attrName === 'content') { + attrValue = attrValue.replace(/\s+/g, '').replace(/[0-9]+\.[0-9]+/g, function(numString) { + // "0.90000" -> "0.9" + // "1.0" -> "1" + // "1.0001" -> "1.0001" (unchanged) + return (+numString).toString(); + }); + } + else if (attrValue && options.customAttrCollapse && options.customAttrCollapse.test(attrName)) { + attrValue = attrValue.replace(/\n+|\r+|\s{2,}/g, ''); + } + else if (tag === 'script' && attrName === 'type') { + attrValue = trimWhitespace(attrValue.replace(/\s*;\s*/g, ';')); + } + else if (isMediaQuery(tag, attrs, attrName)) { + attrValue = trimWhitespace(attrValue); + return unwrapMediaQuery(options.minifyCSS(wrapMediaQuery(attrValue))); + } + return attrValue; +} + +function isMetaViewport(tag, attrs) { + if (tag !== 'meta') { + return false; + } + for (var i = 0, len = attrs.length; i < len; i++) { + if (attrs[i].name === 'name' && attrs[i].value === 'viewport') { + return true; + } + } +} + +// Wrap CSS declarations for CleanCSS > 3.x +// See https://github.com/jakubpawlowicz/clean-css/issues/418 +function wrapInlineCSS(text) { + return '*{' + text + '}'; +} + +function unwrapInlineCSS(text) { + var matches = text.match(/^\*\{([\s\S]*)\}$/); + return matches ? matches[1] : text; +} + +function wrapMediaQuery(text) { + return '@media ' + text + '{a{top:0}}'; +} + +function unwrapMediaQuery(text) { + var matches = text.match(/^@media ([\s\S]*?)\s*{[\s\S]*}$/); + return matches ? matches[1] : text; +} + +function cleanConditionalComment(comment, options) { + return options.processConditionalComments ? comment.replace(/^(\[if\s[^\]]+]>)([\s\S]*?)(<!\[endif])$/, function(match, prefix, text, suffix) { + return prefix + minify(text, options, true) + suffix; + }) : comment; +} + +function processScript(text, options, currentAttrs) { + for (var i = 0, len = currentAttrs.length; i < len; i++) { + if (currentAttrs[i].name.toLowerCase() === 'type' && + options.processScripts.indexOf(currentAttrs[i].value) > -1) { + return minify(text, options); + } + } + return text; +} + +// Tag omission rules from https://html.spec.whatwg.org/multipage/syntax.html#optional-tags +// with the following deviations: +// - retain <body> if followed by <noscript> +// - </rb>, </rt>, </rtc>, </rp> & </tfoot> follow http://www.w3.org/TR/html5/syntax.html#optional-tags +// - retain all tags which are adjacent to non-standard HTML tags +var optionalStartTags = createMapFromString('html,head,body,colgroup,tbody'); +var optionalEndTags = createMapFromString('html,head,body,li,dt,dd,p,rb,rt,rtc,rp,optgroup,option,colgroup,caption,thead,tbody,tfoot,tr,td,th'); +var headerTags = createMapFromString('meta,link,script,style,template,noscript'); +var descriptionTags = createMapFromString('dt,dd'); +var pBlockTags = createMapFromString('address,article,aside,blockquote,details,div,dl,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,hr,main,menu,nav,ol,p,pre,section,table,ul'); +var pInlineTags = createMapFromString('a,audio,del,ins,map,noscript,video'); +var rubyTags = createMapFromString('rb,rt,rtc,rp'); +var rtcTag = createMapFromString('rb,rtc,rp'); +var optionTag = createMapFromString('option,optgroup'); +var tableContentTags = createMapFromString('tbody,tfoot'); +var tableSectionTags = createMapFromString('thead,tbody,tfoot'); +var cellTags = createMapFromString('td,th'); +var topLevelTags = createMapFromString('html,head,body'); +var compactTags = createMapFromString('html,body'); +var looseTags = createMapFromString('head,colgroup,caption'); +var trailingTags = createMapFromString('dt,thead'); +var htmlTags = createMapFromString('a,abbr,acronym,address,applet,area,article,aside,audio,b,base,basefont,bdi,bdo,bgsound,big,blink,blockquote,body,br,button,canvas,caption,center,cite,code,col,colgroup,command,content,data,datalist,dd,del,details,dfn,dialog,dir,div,dl,dt,element,em,embed,fieldset,figcaption,figure,font,footer,form,frame,frameset,h1,h2,h3,h4,h5,h6,head,header,hgroup,hr,html,i,iframe,image,img,input,ins,isindex,kbd,keygen,label,legend,li,link,listing,main,map,mark,marquee,menu,menuitem,meta,meter,multicol,nav,nobr,noembed,noframes,noscript,object,ol,optgroup,option,output,p,param,picture,plaintext,pre,progress,q,rp,rt,rtc,ruby,s,samp,script,section,select,shadow,small,source,spacer,span,strike,strong,style,sub,summary,sup,table,tbody,td,template,textarea,tfoot,th,thead,time,title,tr,track,tt,u,ul,var,video,wbr,xmp'); + +function canRemoveParentTag(optionalStartTag, tag) { + switch (optionalStartTag) { + case 'html': + case 'head': + return true; + case 'body': + return !headerTags(tag); + case 'colgroup': + return tag === 'col'; + case 'tbody': + return tag === 'tr'; + } + return false; +} + +function isStartTagMandatory(optionalEndTag, tag) { + switch (tag) { + case 'colgroup': + return optionalEndTag === 'colgroup'; + case 'tbody': + return tableSectionTags(optionalEndTag); + } + return false; +} + +function canRemovePrecedingTag(optionalEndTag, tag) { + switch (optionalEndTag) { + case 'html': + case 'head': + case 'body': + case 'colgroup': + case 'caption': + return true; + case 'li': + case 'optgroup': + case 'tr': + return tag === optionalEndTag; + case 'dt': + case 'dd': + return descriptionTags(tag); + case 'p': + return pBlockTags(tag); + case 'rb': + case 'rt': + case 'rp': + return rubyTags(tag); + case 'rtc': + return rtcTag(tag); + case 'option': + return optionTag(tag); + case 'thead': + case 'tbody': + return tableContentTags(tag); + case 'tfoot': + return tag === 'tbody'; + case 'td': + case 'th': + return cellTags(tag); + } + return false; +} + +var reEmptyAttribute = new RegExp( + '^(?:class|id|style|title|lang|dir|on(?:focus|blur|change|click|dblclick|mouse(' + + '?:down|up|over|move|out)|key(?:press|down|up)))$'); + +function canDeleteEmptyAttribute(tag, attrName, attrValue, options) { + var isValueEmpty = !attrValue || /^\s*$/.test(attrValue); + if (!isValueEmpty) { + return false; + } + if (typeof options.removeEmptyAttributes === 'function') { + return options.removeEmptyAttributes(attrName, tag); + } + return tag === 'input' && attrName === 'value' || reEmptyAttribute.test(attrName); +} + +function hasAttrName(name, attrs) { + for (var i = attrs.length - 1; i >= 0; i--) { + if (attrs[i].name === name) { + return true; + } + } + return false; +} + +function canRemoveElement(tag, attrs) { + switch (tag) { + case 'textarea': + return false; + case 'audio': + case 'script': + case 'video': + if (hasAttrName('src', attrs)) { + return false; + } + break; + case 'iframe': + if (hasAttrName('src', attrs) || hasAttrName('srcdoc', attrs)) { + return false; + } + break; + case 'object': + if (hasAttrName('data', attrs)) { + return false; + } + break; + case 'applet': + if (hasAttrName('code', attrs)) { + return false; + } + break; + } + return true; +} + +function canCollapseWhitespace(tag) { + return !/^(?:script|style|pre|textarea)$/.test(tag); +} + +function canTrimWhitespace(tag) { + return !/^(?:pre|textarea)$/.test(tag); +} + +function normalizeAttr(attr, attrs, tag, options) { + var attrName = options.caseSensitive ? attr.name : attr.name.toLowerCase(), + attrValue = attr.value; + + if (options.decodeEntities && attrValue) { + attrValue = decode(attrValue, { isAttributeValue: true }); + } + + if (options.removeRedundantAttributes && + isAttributeRedundant(tag, attrName, attrValue, attrs) || + options.removeScriptTypeAttributes && tag === 'script' && + attrName === 'type' && isScriptTypeAttribute(attrValue) || + options.removeStyleLinkTypeAttributes && (tag === 'style' || tag === 'link') && + attrName === 'type' && isStyleLinkTypeAttribute(attrValue)) { + return; + } + + attrValue = cleanAttributeValue(tag, attrName, attrValue, options, attrs); + + if (options.removeEmptyAttributes && + canDeleteEmptyAttribute(tag, attrName, attrValue, options)) { + return; + } + + if (options.decodeEntities && attrValue) { + attrValue = attrValue.replace(/&(#?[0-9a-zA-Z]+;)/g, '&$1'); + } + + return { + attr: attr, + name: attrName, + value: attrValue + }; +} + +function buildAttr(normalized, hasUnarySlash, options, isLast, uidAttr) { + var attrName = normalized.name, + attrValue = normalized.value, + attr = normalized.attr, + attrQuote = attr.quote, + attrFragment, + emittedAttrValue; + + if (typeof attrValue !== 'undefined' && (!options.removeAttributeQuotes || + ~attrValue.indexOf(uidAttr) || !canRemoveAttributeQuotes(attrValue))) { + if (!options.preventAttributesEscaping) { + if (typeof options.quoteCharacter === 'undefined') { + var apos = (attrValue.match(/'/g) || []).length; + var quot = (attrValue.match(/"/g) || []).length; + attrQuote = apos < quot ? '\'' : '"'; + } + else { + attrQuote = options.quoteCharacter === '\'' ? '\'' : '"'; + } + if (attrQuote === '"') { + attrValue = attrValue.replace(/"/g, '"'); + } + else { + attrValue = attrValue.replace(/'/g, '''); + } + } + emittedAttrValue = attrQuote + attrValue + attrQuote; + if (!isLast && !options.removeTagWhitespace) { + emittedAttrValue += ' '; + } + } + // make sure trailing slash is not interpreted as HTML self-closing tag + else if (isLast && !hasUnarySlash && !/\/$/.test(attrValue)) { + emittedAttrValue = attrValue; + } + else { + emittedAttrValue = attrValue + ' '; + } + + if (typeof attrValue === 'undefined' || options.collapseBooleanAttributes && + isBooleanAttribute(attrName.toLowerCase(), attrValue.toLowerCase())) { + attrFragment = attrName; + if (!isLast) { + attrFragment += ' '; + } + } + else { + attrFragment = attrName + attr.customAssign + emittedAttrValue; + } + + return attr.customOpen + attrFragment + attr.customClose; +} + +function identity(value) { + return value; +} + +function processOptions(options) { + ['html5', 'includeAutoGeneratedTags'].forEach(function(key) { + if (!(key in options)) { + options[key] = true; + } + }); + + if (typeof options.log !== 'function') { + options.log = identity; + } + + var defaultTesters = ['canCollapseWhitespace', 'canTrimWhitespace']; + for (var i = 0, len = defaultTesters.length; i < len; i++) { + if (!options[defaultTesters[i]]) { + options[defaultTesters[i]] = function() { + return false; + }; + } + } + + if (!('ignoreCustomComments' in options)) { + options.ignoreCustomComments = [/^!/]; + } + + if (!('ignoreCustomFragments' in options)) { + options.ignoreCustomFragments = [ + /<%[\s\S]*?%>/, + /<\?[\s\S]*?\?>/ + ]; + } + + if (!options.minifyURLs) { + options.minifyURLs = identity; + } + if (typeof options.minifyURLs !== 'function') { + var minifyURLs = options.minifyURLs; + if (typeof minifyURLs === 'string') { + minifyURLs = { site: minifyURLs }; + } + else if (typeof minifyURLs !== 'object') { + minifyURLs = {}; + } + options.minifyURLs = function(text) { + try { + return RelateUrl.relate(text, minifyURLs); + } + catch (err) { + options.log(err); + return text; + } + }; + } + + if (!options.minifyJS) { + options.minifyJS = identity; + } + if (typeof options.minifyJS !== 'function') { + var minifyJS = options.minifyJS; + if (typeof minifyJS !== 'object') { + minifyJS = {}; + } + minifyJS.fromString = true; + (minifyJS.parse || (minifyJS.parse = {})).bare_returns = false; + options.minifyJS = function(text, inline) { + var start = text.match(/^\s*<!--.*/); + var code = start ? text.slice(start[0].length).replace(/\n\s*-->\s*$/, '') : text; + try { + minifyJS.parse.bare_returns = inline; + code = UglifyJS.minify(code, minifyJS).code; + if (/;$/.test(code)) { + code = code.slice(0, -1); + } + return code; + } + catch (err) { + options.log(err); + return text; + } + }; + } + + if (!options.minifyCSS) { + options.minifyCSS = identity; + } + if (typeof options.minifyCSS !== 'function') { + var minifyCSS = options.minifyCSS; + if (typeof minifyCSS !== 'object') { + minifyCSS = {}; + } + options.minifyCSS = function(text) { + text = text.replace(/(url\s*\(\s*)("|'|)(.*?)\2(\s*\))/ig, function(match, prefix, quote, url, suffix) { + return prefix + quote + options.minifyURLs(url) + quote + suffix; + }); + try { + return new CleanCSS(minifyCSS).minify(text).styles; + } + catch (err) { + options.log(err); + return text; + } + }; + } +} + +function uniqueId(value) { + var id; + do { + id = Math.random().toString(36).replace(/^0\.[0-9]*/, ''); + } while (~value.indexOf(id)); + return id; +} + +var specialContentTags = createMapFromString('script,style'); + +function createSortFns(value, options, uidIgnore, uidAttr) { + var attrChains = options.sortAttributes && Object.create(null); + var classChain = options.sortClassName && new TokenChain(); + + function attrNames(attrs) { + return attrs.map(function(attr) { + return options.caseSensitive ? attr.name : attr.name.toLowerCase(); + }); + } + + function shouldSkipUID(token, uid) { + return !uid || token.indexOf(uid) === -1; + } + + function shouldSkipUIDs(token) { + return shouldSkipUID(token, uidIgnore) && shouldSkipUID(token, uidAttr); + } + + function scan(input) { + var currentTag, currentType; + new HTMLParser(input, { + start: function(tag, attrs) { + if (attrChains) { + if (!attrChains[tag]) { + attrChains[tag] = new TokenChain(); + } + attrChains[tag].add(attrNames(attrs).filter(shouldSkipUIDs)); + } + for (var i = 0, len = attrs.length; i < len; i++) { + var attr = attrs[i]; + if (classChain && (options.caseSensitive ? attr.name : attr.name.toLowerCase()) === 'class') { + classChain.add(trimWhitespace(attr.value).split(/\s+/).filter(shouldSkipUIDs)); + } + else if (options.processScripts && attr.name.toLowerCase() === 'type') { + currentTag = tag; + currentType = attr.value; + } + } + }, + end: function() { + currentTag = ''; + }, + chars: function(text) { + if (options.processScripts && specialContentTags(currentTag) && + options.processScripts.indexOf(currentType) > -1) { + scan(text); + } + } + }); + } + + var log = options.log; + options.log = null; + options.sortAttributes = false; + options.sortClassName = false; + scan(minify(value, options)); + options.log = log; + if (attrChains) { + var attrSorters = Object.create(null); + for (var tag in attrChains) { + attrSorters[tag] = attrChains[tag].createSorter(); + } + options.sortAttributes = function(tag, attrs) { + var sorter = attrSorters[tag]; + if (sorter) { + var attrMap = Object.create(null); + var names = attrNames(attrs); + names.forEach(function(name, index) { + (attrMap[name] || (attrMap[name] = [])).push(attrs[index]); + }); + sorter.sort(names).forEach(function(name, index) { + attrs[index] = attrMap[name].shift(); + }); + } + }; + } + if (classChain) { + var sorter = classChain.createSorter(); + options.sortClassName = function(value) { + return sorter.sort(value.split(/\s+/)).join(' '); + }; + } +} + +function minify(value, options, partialMarkup) { + options = options || {}; + var optionsStack = []; + processOptions(options); + if (options.collapseWhitespace) { + value = collapseWhitespace(value, options, true, true); + } + + var buffer = [], + charsPrevTag, + currentChars = '', + hasChars, + currentTag = '', + currentAttrs = [], + stackNoTrimWhitespace = [], + stackNoCollapseWhitespace = [], + optionalStartTag = '', + optionalEndTag = '', + t = Date.now(), + ignoredMarkupChunks = [], + ignoredCustomMarkupChunks = [], + uidIgnore, + uidAttr, + uidPattern; + + // temporarily replace ignored chunks with comments, + // so that we don't have to worry what's there. + // for all we care there might be + // completely-horribly-broken-alien-non-html-emoj-cthulhu-filled content + value = value.replace(/<!-- htmlmin:ignore -->([\s\S]*?)<!-- htmlmin:ignore -->/g, function(match, group1) { + if (!uidIgnore) { + uidIgnore = uniqueId(value); + var pattern = new RegExp('^' + uidIgnore + '([0-9]+)$'); + if (options.ignoreCustomComments) { + options.ignoreCustomComments.push(pattern); + } + else { + options.ignoreCustomComments = [pattern]; + } + } + var token = '<!--' + uidIgnore + ignoredMarkupChunks.length + '-->'; + ignoredMarkupChunks.push(group1); + return token; + }); + + function escapeFragments(text) { + return text.replace(uidPattern, function(match, prefix, index) { + var chunks = ignoredCustomMarkupChunks[+index]; + return chunks[1] + uidAttr + index + chunks[2]; + }); + } + + var customFragments = options.ignoreCustomFragments.map(function(re) { + return re.source; + }); + if (customFragments.length) { + var reCustomIgnore = new RegExp('\\s*(?:' + customFragments.join('|') + ')+\\s*', 'g'); + // temporarily replace custom ignored fragments with unique attributes + value = value.replace(reCustomIgnore, function(match) { + if (!uidAttr) { + uidAttr = uniqueId(value); + uidPattern = new RegExp('(\\s*)' + uidAttr + '([0-9]+)(\\s*)', 'g'); + var minifyCSS = options.minifyCSS; + if (minifyCSS) { + options.minifyCSS = function(text) { + return minifyCSS(escapeFragments(text)); + }; + } + var minifyJS = options.minifyJS; + if (minifyJS) { + options.minifyJS = function(text, inline) { + return minifyJS(escapeFragments(text), inline); + }; + } + } + var token = uidAttr + ignoredCustomMarkupChunks.length; + ignoredCustomMarkupChunks.push(/^(\s*)[\s\S]*?(\s*)$/.exec(match)); + return '\t' + token + '\t'; + }); + } + + if (options.sortAttributes && typeof options.sortAttributes !== 'function' || + options.sortClassName && typeof options.sortClassName !== 'function') { + createSortFns(value, options, uidIgnore, uidAttr); + } + + function _canCollapseWhitespace(tag, attrs) { + return canCollapseWhitespace(tag) || options.canCollapseWhitespace(tag, attrs); + } + + function _canTrimWhitespace(tag, attrs) { + return canTrimWhitespace(tag) || options.canTrimWhitespace(tag, attrs); + } + + function removeStartTag() { + var index = buffer.length - 1; + while (index > 0 && !/^<[^/!]/.test(buffer[index])) { + index--; + } + buffer.length = Math.max(0, index); + } + + function removeEndTag() { + var index = buffer.length - 1; + while (index > 0 && !/^<\//.test(buffer[index])) { + index--; + } + buffer.length = Math.max(0, index); + } + + // look for trailing whitespaces, bypass any inline tags + function trimTrailingWhitespace(index, nextTag) { + for (var endTag = null; index >= 0 && _canTrimWhitespace(endTag); index--) { + var str = buffer[index]; + var match = str.match(/^<\/([\w:-]+)>$/); + if (match) { + endTag = match[1]; + } + else if (/>$/.test(str) || (buffer[index] = collapseWhitespaceSmart(str, null, nextTag, options))) { + break; + } + } + } + + // look for trailing whitespaces from previously processed text + // which may not be trimmed due to a following comment or an empty + // element which has now been removed + function squashTrailingWhitespace(nextTag) { + var charsIndex = buffer.length - 1; + if (buffer.length > 1) { + var item = buffer[buffer.length - 1]; + if (/^(?:<!|$)/.test(item) && item.indexOf(uidIgnore) === -1) { + charsIndex--; + } + } + trimTrailingWhitespace(charsIndex, nextTag); + } + + new HTMLParser(value, { + partialMarkup: partialMarkup, + html5: options.html5, + + start: function(tag, attrs, unary, unarySlash, autoGenerated) { + var lowerTag = tag.toLowerCase(); + + if (lowerTag === 'svg') { + optionsStack.push(options); + var nextOptions = {}; + for (var key in options) { + nextOptions[key] = options[key]; + } + nextOptions.keepClosingSlash = true; + nextOptions.caseSensitive = true; + options = nextOptions; + } + + tag = options.caseSensitive ? tag : lowerTag; + + currentTag = tag; + charsPrevTag = tag; + if (!inlineTextTags(tag)) { + currentChars = ''; + } + hasChars = false; + currentAttrs = attrs; + + var optional = options.removeOptionalTags; + if (optional) { + var htmlTag = htmlTags(tag); + // <html> may be omitted if first thing inside is not comment + // <head> may be omitted if first thing inside is an element + // <body> may be omitted if first thing inside is not space, comment, <meta>, <link>, <script>, <style> or <template> + // <colgroup> may be omitted if first thing inside is <col> + // <tbody> may be omitted if first thing inside is <tr> + if (htmlTag && canRemoveParentTag(optionalStartTag, tag)) { + removeStartTag(); + } + optionalStartTag = ''; + // end-tag-followed-by-start-tag omission rules + if (htmlTag && canRemovePrecedingTag(optionalEndTag, tag)) { + removeEndTag(); + // <colgroup> cannot be omitted if preceding </colgroup> is omitted + // <tbody> cannot be omitted if preceding </tbody>, </thead> or </tfoot> is omitted + optional = !isStartTagMandatory(optionalEndTag, tag); + } + optionalEndTag = ''; + } + + // set whitespace flags for nested tags (eg. <code> within a <pre>) + if (options.collapseWhitespace) { + if (!stackNoTrimWhitespace.length) { + squashTrailingWhitespace(tag); + } + if (!_canTrimWhitespace(tag, attrs)) { + stackNoTrimWhitespace.push(tag); + } + if (!_canCollapseWhitespace(tag, attrs)) { + stackNoCollapseWhitespace.push(tag); + } + } + + var openTag = '<' + tag; + var hasUnarySlash = unarySlash && options.keepClosingSlash; + + buffer.push(openTag); + + if (options.sortAttributes) { + options.sortAttributes(tag, attrs); + } + + var parts = []; + for (var i = attrs.length, isLast = true; --i >= 0;) { + var normalized = normalizeAttr(attrs[i], attrs, tag, options); + if (normalized) { + parts.unshift(buildAttr(normalized, hasUnarySlash, options, isLast, uidAttr)); + isLast = false; + } + } + if (parts.length > 0) { + buffer.push(' '); + buffer.push.apply(buffer, parts); + } + // start tag must never be omitted if it has any attributes + else if (optional && optionalStartTags(tag)) { + optionalStartTag = tag; + } + + buffer.push(buffer.pop() + (hasUnarySlash ? '/' : '') + '>'); + + if (autoGenerated && !options.includeAutoGeneratedTags) { + removeStartTag(); + optionalStartTag = ''; + } + }, + end: function(tag, attrs, autoGenerated) { + var lowerTag = tag.toLowerCase(); + if (lowerTag === 'svg') { + options = optionsStack.pop(); + } + tag = options.caseSensitive ? tag : lowerTag; + + // check if current tag is in a whitespace stack + if (options.collapseWhitespace) { + if (stackNoTrimWhitespace.length) { + if (tag === stackNoTrimWhitespace[stackNoTrimWhitespace.length - 1]) { + stackNoTrimWhitespace.pop(); + } + } + else { + squashTrailingWhitespace('/' + tag); + } + if (stackNoCollapseWhitespace.length && + tag === stackNoCollapseWhitespace[stackNoCollapseWhitespace.length - 1]) { + stackNoCollapseWhitespace.pop(); + } + } + + var isElementEmpty = false; + if (tag === currentTag) { + currentTag = ''; + isElementEmpty = !hasChars; + } + + if (options.removeOptionalTags) { + // <html>, <head> or <body> may be omitted if the element is empty + if (isElementEmpty && topLevelTags(optionalStartTag)) { + removeStartTag(); + } + optionalStartTag = ''; + // </html> or </body> may be omitted if not followed by comment + // </head> may be omitted if not followed by space or comment + // </p> may be omitted if no more content in non-</a> parent + // except for </dt> or </thead>, end tags may be omitted if no more content in parent element + if (htmlTags(tag) && optionalEndTag && !trailingTags(optionalEndTag) && (optionalEndTag !== 'p' || !pInlineTags(tag))) { + removeEndTag(); + } + optionalEndTag = optionalEndTags(tag) ? tag : ''; + } + + if (options.removeEmptyElements && isElementEmpty && canRemoveElement(tag, attrs)) { + // remove last "element" from buffer + removeStartTag(); + optionalStartTag = ''; + optionalEndTag = ''; + } + else { + if (autoGenerated && !options.includeAutoGeneratedTags) { + optionalEndTag = ''; + } + else { + buffer.push('</' + tag + '>'); + } + charsPrevTag = '/' + tag; + if (!inlineTags(tag)) { + currentChars = ''; + } + else if (isElementEmpty) { + currentChars += '|'; + } + } + }, + chars: function(text, prevTag, nextTag) { + prevTag = prevTag === '' ? 'comment' : prevTag; + nextTag = nextTag === '' ? 'comment' : nextTag; + if (options.decodeEntities && text && !specialContentTags(currentTag)) { + text = decode(text); + } + if (options.collapseWhitespace) { + if (!stackNoTrimWhitespace.length) { + if (prevTag === 'comment') { + var prevComment = buffer[buffer.length - 1]; + if (prevComment.indexOf(uidIgnore) === -1) { + if (!prevComment) { + prevTag = charsPrevTag; + } + if (buffer.length > 1 && (!prevComment || !options.conservativeCollapse && / $/.test(currentChars))) { + var charsIndex = buffer.length - 2; + buffer[charsIndex] = buffer[charsIndex].replace(/\s+$/, function(trailingSpaces) { + text = trailingSpaces + text; + return ''; + }); + } + } + } + if (prevTag) { + if (prevTag === '/nobr' || prevTag === 'wbr') { + if (/^\s/.test(text)) { + var tagIndex = buffer.length - 1; + while (tagIndex > 0 && buffer[tagIndex].lastIndexOf('<' + prevTag) !== 0) { + tagIndex--; + } + trimTrailingWhitespace(tagIndex - 1, 'br'); + } + } + else if (inlineTextTags(prevTag.charAt(0) === '/' ? prevTag.slice(1) : prevTag)) { + text = collapseWhitespace(text, options, /(?:^|\s)$/.test(currentChars)); + } + } + if (prevTag || nextTag) { + text = collapseWhitespaceSmart(text, prevTag, nextTag, options); + } + else { + text = collapseWhitespace(text, options, true, true); + } + if (!text && /\s$/.test(currentChars) && prevTag && prevTag.charAt(0) === '/') { + trimTrailingWhitespace(buffer.length - 1, nextTag); + } + } + if (!stackNoCollapseWhitespace.length && nextTag !== 'html' && !(prevTag && nextTag)) { + text = collapseWhitespace(text, options, false, false, true); + } + } + if (options.processScripts && specialContentTags(currentTag)) { + text = processScript(text, options, currentAttrs); + } + if (isExecutableScript(currentTag, currentAttrs)) { + text = options.minifyJS(text); + } + if (isStyleSheet(currentTag, currentAttrs)) { + text = options.minifyCSS(text); + } + if (options.removeOptionalTags && text) { + // <html> may be omitted if first thing inside is not comment + // <body> may be omitted if first thing inside is not space, comment, <meta>, <link>, <script>, <style> or <template> + if (optionalStartTag === 'html' || optionalStartTag === 'body' && !/^\s/.test(text)) { + removeStartTag(); + } + optionalStartTag = ''; + // </html> or </body> may be omitted if not followed by comment + // </head>, </colgroup> or </caption> may be omitted if not followed by space or comment + if (compactTags(optionalEndTag) || looseTags(optionalEndTag) && !/^\s/.test(text)) { + removeEndTag(); + } + optionalEndTag = ''; + } + charsPrevTag = /^\s*$/.test(text) ? prevTag : 'comment'; + if (options.decodeEntities && text && !specialContentTags(currentTag)) { + // semi-colon can be omitted + // https://mathiasbynens.be/notes/ambiguous-ampersands + text = text.replace(/&(#?[0-9a-zA-Z]+;)/g, '&$1').replace(/</g, '<'); + } + currentChars += text; + if (text) { + hasChars = true; + } + buffer.push(text); + }, + comment: function(text, nonStandard) { + var prefix = nonStandard ? '<!' : '<!--'; + var suffix = nonStandard ? '>' : '-->'; + if (isConditionalComment(text)) { + text = prefix + cleanConditionalComment(text, options) + suffix; + } + else if (options.removeComments) { + if (isIgnoredComment(text, options)) { + text = '<!--' + text + '-->'; + } + else { + text = ''; + } + } + else { + text = prefix + text + suffix; + } + if (options.removeOptionalTags && text) { + // preceding comments suppress tag omissions + optionalStartTag = ''; + optionalEndTag = ''; + } + buffer.push(text); + }, + doctype: function(doctype) { + buffer.push(options.useShortDoctype ? '<!DOCTYPE html>' : collapseWhitespaceAll(doctype)); + }, + customAttrAssign: options.customAttrAssign, + customAttrSurround: options.customAttrSurround + }); + + if (options.removeOptionalTags) { + // <html> may be omitted if first thing inside is not comment + // <head> or <body> may be omitted if empty + if (topLevelTags(optionalStartTag)) { + removeStartTag(); + } + // except for </dt> or </thead>, end tags may be omitted if no more content in parent element + if (optionalEndTag && !trailingTags(optionalEndTag)) { + removeEndTag(); + } + } + if (options.collapseWhitespace) { + squashTrailingWhitespace('br'); + } + + var str = joinResultSegments(buffer, options); + + if (uidPattern) { + str = str.replace(uidPattern, function(match, prefix, index, suffix) { + var chunk = ignoredCustomMarkupChunks[+index][0]; + if (options.collapseWhitespace) { + if (prefix !== '\t') { + chunk = prefix + chunk; + } + if (suffix !== '\t') { + chunk += suffix; + } + return collapseWhitespace(chunk, { + preserveLineBreaks: options.preserveLineBreaks, + conservativeCollapse: !options.trimCustomFragments + }, /^\s/.test(chunk), /\s$/.test(chunk)); + } + return chunk; + }); + } + if (uidIgnore) { + str = str.replace(new RegExp('<!--' + uidIgnore + '([0-9]+)-->', 'g'), function(match, index) { + return ignoredMarkupChunks[+index]; + }); + } + + options.log('minified in: ' + (Date.now() - t) + 'ms'); + return str; +} + +function joinResultSegments(results, options) { + var str; + var maxLineLength = options.maxLineLength; + if (maxLineLength) { + var token; + var lines = []; + var line = ''; + for (var i = 0, len = results.length; i < len; i++) { + token = results[i]; + if (line.length + token.length < maxLineLength) { + line += token; + } + else { + lines.push(line.replace(/^\n/, '')); + line = token; + } + } + lines.push(line); + + str = lines.join('\n'); + } + else { + str = results.join(''); + } + return options.collapseWhitespace ? collapseWhitespace(str, options, true, true) : str; +} + +exports.minify = function(value, options) { + return minify(value, options); +}; diff --git a/node_modules/html-minifier/src/htmlparser.js b/node_modules/html-minifier/src/htmlparser.js new file mode 100644 index 000000000..2195347dc --- /dev/null +++ b/node_modules/html-minifier/src/htmlparser.js @@ -0,0 +1,528 @@ +/*! + * HTML Parser By John Resig (ejohn.org) + * Modified by Juriy "kangax" Zaytsev + * Original code by Erik Arvidsson, Mozilla Public License + * http://erik.eae.net/simplehtmlparser/simplehtmlparser.js + */ + +/* + * // Use like so: + * HTMLParser(htmlString, { + * start: function(tag, attrs, unary) {}, + * end: function(tag) {}, + * chars: function(text) {}, + * comment: function(text) {} + * }); + * + * // or to get an XML string: + * HTMLtoXML(htmlString); + * + * // or to get an XML DOM Document + * HTMLtoDOM(htmlString); + * + * // or to inject into an existing document/DOM node + * HTMLtoDOM(htmlString, document); + * HTMLtoDOM(htmlString, document.body); + * + */ + + /* global ActiveXObject, DOMDocument */ + +'use strict'; + +var createMapFromString = require('./utils').createMapFromString; + +function makeMap(values) { + return createMapFromString(values, true); +} + +// Regular Expressions for parsing tags and attributes +var singleAttrIdentifier = /([^\s"'<>/=]+)/, + singleAttrAssigns = [/=/], + singleAttrValues = [ + // attr value double quotes + /"([^"]*)"+/.source, + // attr value, single quotes + /'([^']*)'+/.source, + // attr value, no quotes + /([^ \t\n\f\r"'`=<>]+)/.source + ], + // https://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-QName + qnameCapture = (function() { + var ncname = require('ncname').source.slice(1, -1); + return '((?:' + ncname + '\\:)?' + ncname + ')'; + })(), + startTagOpen = new RegExp('^<' + qnameCapture), + startTagClose = /^\s*(\/?)>/, + endTag = new RegExp('^<\\/' + qnameCapture + '[^>]*>'), + doctype = /^<!DOCTYPE [^>]+>/i; + +var IS_REGEX_CAPTURING_BROKEN = false; +'x'.replace(/x(.)?/g, function(m, g) { + IS_REGEX_CAPTURING_BROKEN = g === ''; +}); + +// Empty Elements +var empty = makeMap('area,base,basefont,br,col,embed,frame,hr,img,input,isindex,keygen,link,meta,param,source,track,wbr'); + +// Inline Elements +var inline = makeMap('a,abbr,acronym,applet,b,basefont,bdo,big,br,button,cite,code,del,dfn,em,font,i,iframe,img,input,ins,kbd,label,map,noscript,object,q,s,samp,script,select,small,span,strike,strong,sub,sup,svg,textarea,tt,u,var'); + +// Elements that you can, intentionally, leave open +// (and which close themselves) +var closeSelf = makeMap('colgroup,dd,dt,li,options,p,td,tfoot,th,thead,tr,source'); + +// Attributes that have their values filled in disabled='disabled' +var fillAttrs = makeMap('checked,compact,declare,defer,disabled,ismap,multiple,nohref,noresize,noshade,nowrap,readonly,selected'); + +// Special Elements (can contain anything) +var special = makeMap('script,style'); + +// HTML5 tags https://html.spec.whatwg.org/multipage/indices.html#elements-3 +// Phrasing Content https://html.spec.whatwg.org/multipage/dom.html#phrasing-content +var nonPhrasing = makeMap('address,article,aside,base,blockquote,body,caption,col,colgroup,dd,details,dialog,div,dl,dt,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,head,header,hgroup,hr,html,legend,li,menuitem,meta,optgroup,option,param,rp,rt,source,style,summary,tbody,td,tfoot,th,thead,title,tr,track'); + +var reCache = {}; + +function attrForHandler(handler) { + var pattern = singleAttrIdentifier.source + + '(?:\\s*(' + joinSingleAttrAssigns(handler) + ')' + + '[ \\t\\n\\f\\r]*(?:' + singleAttrValues.join('|') + '))?'; + if (handler.customAttrSurround) { + var attrClauses = []; + for (var i = handler.customAttrSurround.length - 1; i >= 0; i--) { + attrClauses[i] = '(?:' + + '(' + handler.customAttrSurround[i][0].source + ')\\s*' + + pattern + + '\\s*(' + handler.customAttrSurround[i][1].source + ')' + + ')'; + } + attrClauses.push('(?:' + pattern + ')'); + pattern = '(?:' + attrClauses.join('|') + ')'; + } + return new RegExp('^\\s*' + pattern); +} + +function joinSingleAttrAssigns(handler) { + return singleAttrAssigns.concat( + handler.customAttrAssign || [] + ).map(function(assign) { + return '(?:' + assign.source + ')'; + }).join('|'); +} + +function HTMLParser(html, handler) { + var stack = [], lastTag; + var attribute = attrForHandler(handler); + var last, prevTag, nextTag; + while (html) { + last = html; + // Make sure we're not in a script or style element + if (!lastTag || !special(lastTag)) { + var textEnd = html.indexOf('<'); + if (textEnd === 0) { + // Comment: + if (/^<!--/.test(html)) { + var commentEnd = html.indexOf('-->'); + + if (commentEnd >= 0) { + if (handler.comment) { + handler.comment(html.substring(4, commentEnd)); + } + html = html.substring(commentEnd + 3); + prevTag = ''; + continue; + } + } + + // http://en.wikipedia.org/wiki/Conditional_comment#Downlevel-revealed_conditional_comment + if (/^<!\[/.test(html)) { + var conditionalEnd = html.indexOf(']>'); + + if (conditionalEnd >= 0) { + if (handler.comment) { + handler.comment(html.substring(2, conditionalEnd + 1), true /* non-standard */); + } + html = html.substring(conditionalEnd + 2); + prevTag = ''; + continue; + } + } + + // Doctype: + var doctypeMatch = html.match(doctype); + if (doctypeMatch) { + if (handler.doctype) { + handler.doctype(doctypeMatch[0]); + } + html = html.substring(doctypeMatch[0].length); + prevTag = ''; + continue; + } + + // End tag: + var endTagMatch = html.match(endTag); + if (endTagMatch) { + html = html.substring(endTagMatch[0].length); + endTagMatch[0].replace(endTag, parseEndTag); + prevTag = '/' + endTagMatch[1].toLowerCase(); + continue; + } + + // Start tag: + var startTagMatch = parseStartTag(html); + if (startTagMatch) { + html = startTagMatch.rest; + handleStartTag(startTagMatch); + prevTag = startTagMatch.tagName.toLowerCase(); + continue; + } + } + + var text; + if (textEnd >= 0) { + text = html.substring(0, textEnd); + html = html.substring(textEnd); + } + else { + text = html; + html = ''; + } + + // next tag + var nextTagMatch = parseStartTag(html); + if (nextTagMatch) { + nextTag = nextTagMatch.tagName; + } + else { + nextTagMatch = html.match(endTag); + if (nextTagMatch) { + nextTag = '/' + nextTagMatch[1]; + } + else { + nextTag = ''; + } + } + + if (handler.chars) { + handler.chars(text, prevTag, nextTag); + } + prevTag = ''; + + } + else { + var stackedTag = lastTag.toLowerCase(); + var reStackedTag = reCache[stackedTag] || (reCache[stackedTag] = new RegExp('([\\s\\S]*?)</' + stackedTag + '[^>]*>', 'i')); + + html = html.replace(reStackedTag, function(all, text) { + if (stackedTag !== 'script' && stackedTag !== 'style' && stackedTag !== 'noscript') { + text = text + .replace(/<!--([\s\S]*?)-->/g, '$1') + .replace(/<!\[CDATA\[([\s\S]*?)]]>/g, '$1'); + } + + if (handler.chars) { + handler.chars(text); + } + + return ''; + }); + + parseEndTag('</' + stackedTag + '>', stackedTag); + } + + if (html === last) { + throw new Error('Parse Error: ' + html); + } + } + + if (!handler.partialMarkup) { + // Clean up any remaining tags + parseEndTag(); + } + + function parseStartTag(input) { + var start = input.match(startTagOpen); + if (start) { + var match = { + tagName: start[1], + attrs: [] + }; + input = input.slice(start[0].length); + var end, attr; + while (!(end = input.match(startTagClose)) && (attr = input.match(attribute))) { + input = input.slice(attr[0].length); + match.attrs.push(attr); + } + if (end) { + match.unarySlash = end[1]; + match.rest = input.slice(end[0].length); + return match; + } + } + } + + function handleStartTag(match) { + var tagName = match.tagName; + var unarySlash = match.unarySlash; + + if (handler.html5 && lastTag === 'p' && nonPhrasing(tagName)) { + parseEndTag('', lastTag); + } + + if (!handler.html5) { + while (lastTag && inline(lastTag)) { + parseEndTag('', lastTag); + } + } + + if (closeSelf(tagName) && lastTag === tagName) { + parseEndTag('', tagName); + } + + var unary = empty(tagName) || tagName === 'html' && lastTag === 'head' || !!unarySlash; + + var attrs = match.attrs.map(function(args) { + var name, value, customOpen, customClose, customAssign, quote; + var ncp = 7; // number of captured parts, scalar + + // hackish work around FF bug https://bugzilla.mozilla.org/show_bug.cgi?id=369778 + if (IS_REGEX_CAPTURING_BROKEN && args[0].indexOf('""') === -1) { + if (args[3] === '') { delete args[3]; } + if (args[4] === '') { delete args[4]; } + if (args[5] === '') { delete args[5]; } + } + + function populate(index) { + customAssign = args[index]; + value = args[index + 1]; + if (typeof value !== 'undefined') { + return '"'; + } + value = args[index + 2]; + if (typeof value !== 'undefined') { + return '\''; + } + value = args[index + 3]; + if (typeof value === 'undefined' && fillAttrs(name)) { + value = name; + } + return ''; + } + + var j = 1; + if (handler.customAttrSurround) { + for (var i = 0, l = handler.customAttrSurround.length; i < l; i++, j += ncp) { + name = args[j + 1]; + if (name) { + quote = populate(j + 2); + customOpen = args[j]; + customClose = args[j + 6]; + break; + } + } + } + + if (!name && (name = args[j])) { + quote = populate(j + 1); + } + + return { + name: name, + value: value, + customAssign: customAssign || '=', + customOpen: customOpen || '', + customClose: customClose || '', + quote: quote || '' + }; + }); + + if (!unary) { + stack.push({ tag: tagName, attrs: attrs }); + lastTag = tagName; + unarySlash = ''; + } + + if (handler.start) { + handler.start(tagName, attrs, unary, unarySlash); + } + } + + function parseEndTag(tag, tagName) { + var pos; + + // Find the closest opened tag of the same type + if (tagName) { + var needle = tagName.toLowerCase(); + for (pos = stack.length - 1; pos >= 0; pos--) { + if (stack[pos].tag.toLowerCase() === needle) { + break; + } + } + } + // If no tag name is provided, clean shop + else { + pos = 0; + } + + if (pos >= 0) { + // Close all the open elements, up the stack + for (var i = stack.length - 1; i >= pos; i--) { + if (handler.end) { + handler.end(stack[i].tag, stack[i].attrs, i > pos || !tag); + } + } + + // Remove the open elements from the stack + stack.length = pos; + lastTag = pos && stack[pos - 1].tag; + } + else if (tagName.toLowerCase() === 'br') { + if (handler.start) { + handler.start(tagName, [], true, ''); + } + } + else if (tagName.toLowerCase() === 'p') { + if (handler.start) { + handler.start(tagName, [], false, '', true); + } + if (handler.end) { + handler.end(tagName, []); + } + } + } +} + +exports.HTMLParser = HTMLParser; +exports.HTMLtoXML = function(html) { + var results = ''; + + new HTMLParser(html, { + start: function(tag, attrs, unary) { + results += '<' + tag; + + for (var i = 0, len = attrs.length; i < len; i++) { + results += ' ' + attrs[i].name + '="' + (attrs[i].value || '').replace(/"/g, '"') + '"'; + } + + results += (unary ? '/' : '') + '>'; + }, + end: function(tag) { + results += '</' + tag + '>'; + }, + chars: function(text) { + results += text; + }, + comment: function(text) { + results += '<!--' + text + '-->'; + }, + ignore: function(text) { + results += text; + } + }); + + return results; +}; + +exports.HTMLtoDOM = function(html, doc) { + // There can be only one of these elements + var one = { + html: true, + head: true, + body: true, + title: true + }; + + // Enforce a structure for the document + var structure = { + link: 'head', + base: 'head' + }; + + if (doc) { + doc = doc.ownerDocument || doc.getOwnerDocument && doc.getOwnerDocument() || doc; + } + else if (typeof DOMDocument !== 'undefined') { + doc = new DOMDocument(); + } + else if (typeof document !== 'undefined' && document.implementation && document.implementation.createDocument) { + doc = document.implementation.createDocument('', '', null); + } + else if (typeof ActiveX !== 'undefined') { + doc = new ActiveXObject('Msxml.DOMDocument'); + } + + var elems = [], + documentElement = doc.documentElement || + doc.getDocumentElement && doc.getDocumentElement(); + + // If we're dealing with an empty document then we + // need to pre-populate it with the HTML document structure + if (!documentElement && doc.createElement) { + (function() { + var html = doc.createElement('html'); + var head = doc.createElement('head'); + head.appendChild(doc.createElement('title')); + html.appendChild(head); + html.appendChild(doc.createElement('body')); + doc.appendChild(html); + })(); + } + + // Find all the unique elements + if (doc.getElementsByTagName) { + for (var i in one) { + one[i] = doc.getElementsByTagName(i)[0]; + } + } + + // If we're working with a document, inject contents into + // the body element + var curParentNode = one.body; + + new HTMLParser(html, { + start: function(tagName, attrs, unary) { + // If it's a pre-built element, then we can ignore + // its construction + if (one[tagName]) { + curParentNode = one[tagName]; + return; + } + + var elem = doc.createElement(tagName); + + for (var attr in attrs) { + elem.setAttribute(attrs[attr].name, attrs[attr].value); + } + + if (structure[tagName] && typeof one[structure[tagName]] !== 'boolean') { + one[structure[tagName]].appendChild(elem); + } + else if (curParentNode && curParentNode.appendChild) { + curParentNode.appendChild(elem); + } + + if (!unary) { + elems.push(elem); + curParentNode = elem; + } + }, + end: function(/* tag */) { + elems.length -= 1; + + // Init the new parentNode + curParentNode = elems[elems.length - 1]; + }, + chars: function(text) { + curParentNode.appendChild(doc.createTextNode(text)); + }, + comment: function(/* text */) { + // create comment node + }, + ignore: function(/* text */) { + // What to do here? + } + }); + + return doc; +}; diff --git a/node_modules/html-minifier/src/tokenchain.js b/node_modules/html-minifier/src/tokenchain.js new file mode 100644 index 000000000..f8892d017 --- /dev/null +++ b/node_modules/html-minifier/src/tokenchain.js @@ -0,0 +1,71 @@ +'use strict'; + +function Sorter() { +} + +Sorter.prototype.sort = function(tokens, fromIndex) { + fromIndex = fromIndex || 0; + for (var i = 0, len = this.keys.length; i < len; i++) { + var key = this.keys[i]; + var token = key.slice(1); + var index = tokens.indexOf(token, fromIndex); + if (index !== -1) { + do { + if (index !== fromIndex) { + tokens.splice(index, 1); + tokens.splice(fromIndex, 0, token); + } + fromIndex++; + } while ((index = tokens.indexOf(token, fromIndex)) !== -1); + return this[key].sort(tokens, fromIndex); + } + } + return tokens; +}; + +function TokenChain() { +} + +TokenChain.prototype = { + add: function(tokens) { + var self = this; + tokens.forEach(function(token) { + var key = '$' + token; + if (!self[key]) { + self[key] = []; + self[key].processed = 0; + } + self[key].push(tokens); + }); + }, + createSorter: function() { + var self = this; + var sorter = new Sorter(); + sorter.keys = Object.keys(self).sort(function(j, k) { + var m = self[j].length; + var n = self[k].length; + return m < n ? 1 : m > n ? -1 : j < k ? -1 : j > k ? 1 : 0; + }).filter(function(key) { + if (self[key].processed < self[key].length) { + var token = key.slice(1); + var chain = new TokenChain(); + self[key].forEach(function(tokens) { + var index; + while ((index = tokens.indexOf(token)) !== -1) { + tokens.splice(index, 1); + } + tokens.forEach(function(token) { + self['$' + token].processed++; + }); + chain.add(tokens.slice(0)); + }); + sorter[key] = chain.createSorter(); + return true; + } + return false; + }); + return sorter; + } +}; + +module.exports = TokenChain; diff --git a/node_modules/html-minifier/src/utils.js b/node_modules/html-minifier/src/utils.js new file mode 100644 index 000000000..62b43636d --- /dev/null +++ b/node_modules/html-minifier/src/utils.js @@ -0,0 +1,18 @@ +'use strict'; + +function createMap(values, ignoreCase) { + var map = {}; + values.forEach(function(value) { + map[value] = 1; + }); + return ignoreCase ? function(value) { + return map[value.toLowerCase()] === 1; + } : function(value) { + return map[value] === 1; + }; +} + +exports.createMap = createMap; +exports.createMapFromString = function(values, ignoreCase) { + return createMap(values.split(/,/), ignoreCase); +}; |