"use strict"; var __extends = (this && this.__extends) || (function () { var extendStatics = function (d, b) { extendStatics = Object.setPrototypeOf || ({ __proto__: [] } instanceof Array && function (d, b) { d.__proto__ = b; }) || function (d, b) { for (var p in b) if (b.hasOwnProperty(p)) d[p] = b[p]; }; return extendStatics(d, b); }; return function (d, b) { extendStatics(d, b); function __() { this.constructor = d; } d.prototype = b === null ? Object.create(b) : (__.prototype = b.prototype, new __()); }; })(); var __importDefault = (this && this.__importDefault) || function (mod) { return (mod && mod.__esModule) ? mod : { "default": mod }; }; Object.defineProperty(exports, "__esModule", { value: true }); var Tokenizer_1 = __importDefault(require("./Tokenizer")); var events_1 = require("events"); var formTags = new Set([ "input", "option", "optgroup", "select", "button", "datalist", "textarea" ]); var pTag = new Set(["p"]); var openImpliesClose = { tr: new Set(["tr", "th", "td"]), th: new Set(["th"]), td: new Set(["thead", "th", "td"]), body: new Set(["head", "link", "script"]), li: new Set(["li"]), p: pTag, h1: pTag, h2: pTag, h3: pTag, h4: pTag, h5: pTag, h6: pTag, select: formTags, input: formTags, output: formTags, button: formTags, datalist: formTags, textarea: formTags, option: new Set(["option"]), optgroup: new Set(["optgroup", "option"]), dd: new Set(["dt", "dd"]), dt: new Set(["dt", "dd"]), address: pTag, article: pTag, aside: pTag, blockquote: pTag, details: pTag, div: pTag, dl: pTag, fieldset: pTag, figcaption: pTag, figure: pTag, footer: pTag, form: pTag, header: pTag, hr: pTag, main: pTag, nav: pTag, ol: pTag, pre: pTag, section: pTag, table: pTag, ul: pTag, rt: new Set(["rt", "rp"]), rp: new Set(["rt", "rp"]), tbody: new Set(["thead", "tbody"]), tfoot: new Set(["thead", "tbody"]) }; var voidElements = new Set([ "area", "base", "basefont", "br", "col", "command", "embed", "frame", "hr", "img", "input", "isindex", "keygen", "link", "meta", "param", "source", "track", "wbr" ]); var foreignContextElements = new Set(["math", "svg"]); var htmlIntegrationElements = new Set([ "mi", "mo", "mn", "ms", "mtext", "annotation-xml", "foreignObject", "desc", "title" ]); var reNameEnd = /\s|\//; var Parser = /** @class */ (function (_super) { __extends(Parser, _super); function Parser(cbs, options) { var _this = _super.call(this) || this; _this._tagname = ""; _this._attribname = ""; _this._attribvalue = ""; _this._attribs = null; _this._stack = []; _this._foreignContext = []; _this.startIndex = 0; _this.endIndex = null; // Aliases for backwards compatibility _this.parseChunk = Parser.prototype.write; _this.done = Parser.prototype.end; _this._options = options || {}; _this._cbs = cbs || {}; _this._tagname = ""; _this._attribname = ""; _this._attribvalue = ""; _this._attribs = null; _this._stack = []; _this._foreignContext = []; _this.startIndex = 0; _this.endIndex = null; _this._lowerCaseTagNames = "lowerCaseTags" in _this._options ? !!_this._options.lowerCaseTags : !_this._options.xmlMode; _this._lowerCaseAttributeNames = "lowerCaseAttributeNames" in _this._options ? !!_this._options.lowerCaseAttributeNames : !_this._options.xmlMode; _this._tokenizer = new (_this._options.Tokenizer || Tokenizer_1.default)(_this._options, _this); if (_this._cbs.onparserinit) _this._cbs.onparserinit(_this); return _this; } Parser.prototype._updatePosition = function (initialOffset) { if (this.endIndex === null) { if (this._tokenizer._sectionStart <= initialOffset) { this.startIndex = 0; } else { this.startIndex = this._tokenizer._sectionStart - initialOffset; } } else this.startIndex = this.endIndex + 1; this.endIndex = this._tokenizer.getAbsoluteIndex(); }; //Tokenizer event handlers Parser.prototype.ontext = function (data) { this._updatePosition(1); // @ts-ignore this.endIndex--; if (this._cbs.ontext) this._cbs.ontext(data); }; Parser.prototype.onopentagname = function (name) { if (this._lowerCaseTagNames) { name = name.toLowerCase(); } this._tagname = name; if (!this._options.xmlMode && Object.prototype.hasOwnProperty.call(openImpliesClose, name)) { for (var el = void 0; // @ts-ignore openImpliesClose[name].has((el = this._stack[this._stack.length - 1])); this.onclosetag(el)) ; } if (this._options.xmlMode || !voidElements.has(name)) { this._stack.push(name); if (foreignContextElements.has(name)) { this._foreignContext.push(true); } else if (htmlIntegrationElements.has(name)) { this._foreignContext.push(false); } } if (this._cbs.onopentagname) this._cbs.onopentagname(name); if (this._cbs.onopentag) this._attribs = {}; }; Parser.prototype.onopentagend = function () { this._updatePosition(1); if (this._attribs) { if (this._cbs.onopentag) { this._cbs.onopentag(this._tagname, this._attribs); } this._attribs = null; } if (!this._options.xmlMode && this._cbs.onclosetag && voidElements.has(this._tagname)) { this._cbs.onclosetag(this._tagname); } this._tagname = ""; }; Parser.prototype.onclosetag = function (name) { this._updatePosition(1); if (this._lowerCaseTagNames) { name = name.toLowerCase(); } if (foreignContextElements.has(name) || htmlIntegrationElements.has(name)) { this._foreignContext.pop(); } if (this._stack.length && (this._options.xmlMode || !voidElements.has(name))) { var pos = this._stack.lastIndexOf(name); if (pos !== -1) { if (this._cbs.onclosetag) { pos = this._stack.length - pos; // @ts-ignore while (pos--) this._cbs.onclosetag(this._stack.pop()); } else this._stack.length = pos; } else if (name === "p" && !this._options.xmlMode) { this.onopentagname(name); this._closeCurrentTag(); } } else if (!this._options.xmlMode && (name === "br" || name === "p")) { this.onopentagname(name); this._closeCurrentTag(); } }; Parser.prototype.onselfclosingtag = function () { if (this._options.xmlMode || this._options.recognizeSelfClosing || this._foreignContext[this._foreignContext.length - 1]) { this._closeCurrentTag(); } else { this.onopentagend(); } }; Parser.prototype._closeCurrentTag = function () { var name = this._tagname; this.onopentagend(); //self-closing tags will be on the top of the stack //(cheaper check than in onclosetag) if (this._stack[this._stack.length - 1] === name) { if (this._cbs.onclosetag) { this._cbs.onclosetag(name); } this._stack.pop(); } }; Parser.prototype.onattribname = function (name) { if (this._lowerCaseAttributeNames) { name = name.toLowerCase(); } this._attribname = name; }; Parser.prototype.onattribdata = function (value) { this._attribvalue += value; }; Parser.prototype.onattribend = function () { if (this._cbs.onattribute) this._cbs.onattribute(this._attribname, this._attribvalue); if (this._attribs && !Object.prototype.hasOwnProperty.call(this._attribs, this._attribname)) { this._attribs[this._attribname] = this._attribvalue; } this._attribname = ""; this._attribvalue = ""; }; Parser.prototype._getInstructionName = function (value) { var idx = value.search(reNameEnd); var name = idx < 0 ? value : value.substr(0, idx); if (this._lowerCaseTagNames) { name = name.toLowerCase(); } return name; }; Parser.prototype.ondeclaration = function (value) { if (this._cbs.onprocessinginstruction) { var name_1 = this._getInstructionName(value); this._cbs.onprocessinginstruction("!" + name_1, "!" + value); } }; Parser.prototype.onprocessinginstruction = function (value) { if (this._cbs.onprocessinginstruction) { var name_2 = this._getInstructionName(value); this._cbs.onprocessinginstruction("?" + name_2, "?" + value); } }; Parser.prototype.oncomment = function (value) { this._updatePosition(4); if (this._cbs.oncomment) this._cbs.oncomment(value); if (this._cbs.oncommentend) this._cbs.oncommentend(); }; Parser.prototype.oncdata = function (value) { this._updatePosition(1); if (this._options.xmlMode || this._options.recognizeCDATA) { if (this._cbs.oncdatastart) this._cbs.oncdatastart(); if (this._cbs.ontext) this._cbs.ontext(value); if (this._cbs.oncdataend) this._cbs.oncdataend(); } else { this.oncomment("[CDATA[" + value + "]]"); } }; Parser.prototype.onerror = function (err) { if (this._cbs.onerror) this._cbs.onerror(err); }; Parser.prototype.onend = function () { if (this._cbs.onclosetag) { for (var i = this._stack.length; i > 0; this._cbs.onclosetag(this._stack[--i])) ; } if (this._cbs.onend) this._cbs.onend(); }; //Resets the parser to a blank state, ready to parse a new HTML document Parser.prototype.reset = function () { if (this._cbs.onreset) this._cbs.onreset(); this._tokenizer.reset(); this._tagname = ""; this._attribname = ""; this._attribs = null; this._stack = []; if (this._cbs.onparserinit) this._cbs.onparserinit(this); }; //Parses a complete HTML document and pushes it to the handler Parser.prototype.parseComplete = function (data) { this.reset(); this.end(data); }; Parser.prototype.write = function (chunk) { this._tokenizer.write(chunk); }; Parser.prototype.end = function (chunk) { this._tokenizer.end(chunk); }; Parser.prototype.pause = function () { this._tokenizer.pause(); }; Parser.prototype.resume = function () { this._tokenizer.resume(); }; return Parser; }(events_1.EventEmitter)); exports.Parser = Parser;