wikiparser-node 0.5.0 → 0.6.1

package/lib/text.js

@@ -23,6 +23,68 @@ class AstText extends AstNode {
 		});
 	}
 
+	/** 输出字符串 */
+	toString() {
+		return this.data;
+	}
+
+	/**
+	 * 修改内容
+	 * @param {string} text 新内容
+	 */
+	#setData(text) {
+		text = String(text);
+		const {data} = this,
+			e = new Event('text', {bubbles: true});
+		this.setAttribute('data', text);
+		if (data !== text) {
+			this.dispatchEvent(e, {oldText: data, newText: text});
+		}
+		return this;
+	}
+
+	/**
+	 * 替换字符串
+	 * @param {string} text 替换的字符串
+	 */
+	replaceData(text = '') {
+		this.#setData(text);
+	}
+
+	static errorSyntax = /[{}]+|\[{2,}|\[(?!(?:(?!https?\b)[^[])*\])|(?<=^|\])([^[]*?)\]+|<(?=\s*\/?\w+[\s/>])/giu;
+
+	/**
+	 * Linter
+	 * @param {number} start 起始位置
+	 * @returns {LintError[]}
+	 */
+	lint(start = 0) {
+		const {data} = this,
+			errors = [...data.matchAll(AstText.errorSyntax)];
+		if (errors.length > 0) {
+			const {top, left} = this.getRootNode().posFromIndex(start);
+			return errors.map(({0: error, 1: prefix, index}) => {
+				if (prefix) {
+					index += prefix.length;
+					error = error.slice(prefix.length);
+				}
+				const lines = data.slice(0, index).split('\n'),
+					startLine = lines.length + top - 1,
+					{length} = lines.at(-1),
+					startCol = lines.length > 1 ? length : left + length;
+				return {
+					message: `孤立的"${error[0]}"`,
+					severity: error[0] === '{' || error[0] === '}' ? 'error' : 'warning',
+					startLine,
+					endLine: startLine,
+					startCol,
+					endCol: startCol + error.length,
+				};
+			});
+		}
+		return [];
+	}
+
 	/** 复制 */
 	cloneNode() {
 		return new AstText(this.data);
@@ -43,31 +105,11 @@ class AstText extends AstNode {
 			: super.getAttribute(key);
 	}
 
-	/** 输出字符串 */
-	toString() {
-		return this.data;
-	}
-
 	/** @override */
 	text() {
 		return this.data;
 	}
 
-	/**
-	 * 修改内容
-	 * @param {string} text 新内容
-	 */
-	#setData(text) {
-		text = String(text);
-		const {data} = this,
-			e = new Event('text', {bubbles: true});
-		this.setAttribute('data', text);
-		if (data !== text) {
-			this.dispatchEvent(e, {oldText: data, newText: text});
-		}
-		return this;
-	}
-
 	/**
 	 * 在后方添加字符串
 	 * @param {string} text 添加的字符串
@@ -94,14 +136,6 @@ class AstText extends AstNode {
 		this.#setData(this.data.slice(0, offset) + text + this.data.slice(offset));
 	}
 
-	/**
-	 * 替换字符串
-	 * @param {string} text 替换的字符串
-	 */
-	replaceData(text = '') {
-		this.#setData(text);
-	}
-
 	/**
 	 * 提取子串
 	 * @param {number} offset 起始位置
@@ -118,9 +152,9 @@ class AstText extends AstNode {
 	 * @throws `Error` 没有父节点
 	 */
 	splitText(offset) {
-		if (typeof offset !== 'number') {
+		if (!Number.isInteger(offset)) {
 			this.typeError('splitText', 'Number');
-		} else if (offset > this.length || offset < -this.length || !Number.isInteger(offset)) {
+		} else if (offset > this.length || offset < -this.length) {
 			throw new RangeError(`错误的断开位置！${offset}`);
 		}
 		const {parentNode, data} = this;
@@ -135,40 +169,6 @@ class AstText extends AstNode {
 		parentNode.setAttribute('childNodes', childNodes);
 		return newText;
 	}
-
-	static errorSyntax = /[{}]+|\[{2,}|\[(?!(?:(?!https?\b)[^[])*\])|(?<=^|\])([^[]*?)\]+|<(?=\s*\/?\w+[\s/>])/giu;
-
-	/**
-	 * Linter
-	 * @param {number} start 起始位置
-	 * @returns {LintError[]}
-	 */
-	lint(start = 0) {
-		const {data} = this,
-			errors = [...data.matchAll(AstText.errorSyntax)];
-		if (errors.length > 0) {
-			const {top, left} = this.getRootNode().posFromIndex(start);
-			return errors.map(({0: error, 1: prefix, index}) => {
-				if (prefix) {
-					index += prefix.length;
-					error = error.slice(prefix.length);
-				}
-				const lines = data.slice(0, index).split('\n'),
-					startLine = lines.length + top - 1,
-					{length} = lines.at(-1),
-					startCol = lines.length > 1 ? length : left + length;
-				return {
-					message: `孤立的"${error[0]}"`,
-					severity: error[0] === '{' || error[0] === '}' ? 'error' : 'warning',
-					startLine,
-					endLine: startLine,
-					startCol,
-					endCol: startCol + error.length,
-				};
-			});
-		}
-		return [];
-	}
 }
 
 Parser.classes.AstText = __filename;

package/lib/title.js

@@ -4,13 +4,13 @@ const Parser = require('..');
 
 /** MediaWiki页面标题对象 */
 class Title {
+	valid = true;
+	ns = 0;
 	title = '';
 	main = '';
 	prefix = '';
-	ns = 0;
 	interwiki = '';
 	fragment = '';
-	valid = true;
 
 	/**
 	 * @param {string} title 标题（含或不含命名空间前缀）
@@ -39,24 +39,25 @@ class Title {
 		}
 		this.ns = nsid[namespace.toLowerCase()];
 		const i = title.indexOf('#');
+		let fragment = '';
 		if (i !== -1) {
-			const fragment = title.slice(i + 1).trimEnd();
+			fragment = title.slice(i + 1).trimEnd();
 			if (fragment.includes('%')) {
 				try {
-					this.fragment = decodeURIComponent(fragment);
+					fragment = decodeURIComponent(fragment);
 				} catch {}
 			} else if (fragment.includes('.')) {
 				try {
-					this.fragment = decodeURIComponent(fragment.replaceAll('.', '%'));
+					fragment = decodeURIComponent(fragment.replaceAll('.', '%'));
 				} catch {}
 			}
-			this.fragment ||= fragment;
 			title = title.slice(0, i).trim();
 		}
+		this.valid = Boolean(title || fragment) && !/\0\d+[eh!+-]\x7F|[<>[\]{}|]/u.test(title);
 		this.main = title && `${title[0].toUpperCase()}${title.slice(1)}`;
 		this.prefix = `${namespace}${namespace && ':'}`;
 		this.title = `${iw ? `${this.interwiki}:` : ''}${this.prefix}${this.main.replaceAll(' ', '_')}`;
-		this.valid = Boolean(this.main || this.fragment) && !/\0\d+[eh!+-]\x7F|[<>[\]{}|]/u.test(this.title);
+		this.fragment = fragment;
 	}
 
 	/** @override */

package/mixin/hidden.js

@@ -9,6 +9,8 @@ const Parser = require('..');
  * @returns {T}
  */
 const hidden = Constructor => class extends Constructor {
+	static hidden = true;
+
 	/** 没有可见部分 */
 	text() { // eslint-disable-line class-methods-use-this
 		return '';

package/mixin/sol.js

@@ -18,8 +18,7 @@ const sol = Constructor => class SolToken extends Constructor {
 	#isRoot(includeHeading) {
 		const {parentNode, type} = this;
 		return parentNode?.type === 'root'
-			|| parentNode?.type === 'ext-inner' && parentNode?.name === 'poem'
-				&& (includeHeading || type !== 'heading');
+			|| parentNode?.type === 'ext-inner' && (includeHeading || type !== 'heading' && parentNode.name === 'poem');
 	}
 
 	/**

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "wikiparser-node",
-  "version": "0.5.0",
+  "version": "0.6.1",
   "description": "A Node.js parser for MediaWiki markup with AST",
   "keywords": [
     "mediawiki",
@@ -28,8 +28,9 @@
     "url": "git+https://github.com/bhsd-harry/wikiparser-node.git"
   },
   "scripts": {
-    "test": "node test/test.js",
-    "real": "node test/real.js"
+    "test": "eslint . && node test/test.js",
+    "real": "node test/real.js",
+    "single": "node --prof test/single.js && node --prof-process isolate-0x*-v8.log > test/processed.txt && rm isolate-0x*-v8.log"
   },
   "devDependencies": {
     "@types/node": "^17.0.23",

package/parser/brackets.js

@@ -57,12 +57,18 @@ const parseBrackets = (text, config = Parser.getConfig(), accum = []) => {
 				{length} = accum;
 			lastIndex = curIndex + close.length; // 这不是最终的lastIndex
 			parts.at(-1).push(text.slice(topPos, curIndex));
-			const ch = close.length === 2 ? marks[removeComment(parts[0][0])] ?? 't' : 't'; // 标记{{!}}等
-			let skip = false;
+			let skip = false,
+				ch = 't';
 			if (close.length === 3) {
 				const ArgToken = require('../src/arg');
 				new ArgToken(parts.map(part => part.join('=')), config, accum);
 			} else {
+				const name = removeComment(parts[0][0]);
+				if (name in marks) {
+					ch = marks[name]; // 标记{{!}}等
+				} else if (/^(?:fullurl|canonicalurl|filepath):./iu.test(name)) {
+					ch = 'm';
+				}
 				try {
 					const TranscludeToken = require('../src/transclude');
 					new TranscludeToken(parts[0][0], parts.slice(1), config, accum);

package/parser/externalLinks.js

@@ -11,7 +11,7 @@ const {extUrlChar} = require('../util/string'),
 const parseExternalLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
 	const ExtLinkToken = require('../src/extLink');
 	const regex = new RegExp(
-		`\\[((?:${config.protocol}|//)${extUrlChar})(\\p{Zs}*)([^\\]\x01-\x08\x0A-\x1F\uFFFD]*)\\]`,
+		`\\[((?:${config.protocol}|//)${extUrlChar}|\0\\d+m\x7F)(\\p{Zs}*)([^\\]\x01-\x08\x0A-\x1F\uFFFD]*)\\]`,
 		'giu',
 	);
 	return wikitext.replaceAll(regex, /** @type {function(...string): string} */ (_, url, space, text) => {

package/parser/hrAndDoubleUnderscore.js

@@ -1,8 +1,6 @@
 'use strict';
 
-const Parser = require('..'),
-	AstText = require('../lib/text'),
-	Token = require('../src');
+const Parser = require('..');
 
 /**
  * 解析\<hr\>和状态开关
@@ -10,7 +8,9 @@ const Parser = require('..'),
  * @param {accum} accum
  */
 const parseHrAndDoubleUnderscore = ({firstChild: {data}, type, name}, config = Parser.getConfig(), accum = []) => {
-	const HrToken = require('../src/nowiki/hr'),
+	const AstText = require('../lib/text'),
+		Token = require('../src'),
+		HrToken = require('../src/nowiki/hr'),
 		DoubleUnderscoreToken = require('../src/nowiki/doubleUnderscore');
 	const {doubleUnderscore} = config;
 	if (type !== 'root' && (type !== 'ext-inner' || name !== 'poem')) {

package/parser/links.js

@@ -9,17 +9,17 @@ const Parser = require('..');
  */
 const parseLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
 	const parseQuotes = require('./quotes.js');
-	const regex = /^([^\n<>[\]{}|]+)(?:\|(.*?[^\]]))?\]\](.*)$/su,
-		regexImg = /^([^\n<>[\]{}|]+)\|(.*)$/su,
+	const regex = /^((?:(?!\0\d+!\x7F)[^\n<>[\]{}|])+)(?:(\||\0\d+!\x7F)(.*?[^\]]))?\]\](.*)$/su,
+		regexImg = /^((?:(?!\0\d+!\x7F)[^\n<>[\]{}|])+)(\||\0\d+!\x7F)(.*)$/su,
 		regexExt = new RegExp(`^\\s*(?:${config.protocol})`, 'iu'),
 		bits = wikitext.split('[[');
 	let s = bits.shift();
 	for (let i = 0; i < bits.length; i++) {
-		let mightBeImg, link, text, after;
+		let mightBeImg, link, delimiter, text, after;
 		const x = bits[i],
 			m = regex.exec(x);
 		if (m) {
-			[, link, text, after] = m;
+			[, link, delimiter, text, after] = m;
 			if (after[0] === ']' && text?.includes('[')) {
 				text += ']';
 				after = after.slice(1);
@@ -28,7 +28,7 @@ const parseLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
 			const m2 = regexImg.exec(x);
 			if (m2) {
 				mightBeImg = true;
-				[, link, text] = m2;
+				[, link, delimiter, text] = m2;
 			}
 		}
 		if (link === undefined || regexExt.test(link) || /\0\d+[exhbru]\x7F/u.test(link)) {
@@ -74,7 +74,7 @@ const parseLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
 			}
 			text = parseLinks(text, config, accum);
 			if (!found) {
-				s += `[[${link}|${text}`;
+				s += `[[${link}${delimiter}${text}`;
 				continue;
 			}
 		}
@@ -88,7 +88,7 @@ const parseLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
 				LinkToken = require('../src/link/category');
 			}
 		}
-		new LinkToken(link, text, title, config, accum);
+		new LinkToken(link, text, title, config, accum, delimiter);
 	}
 	return s;
 };

package/parser/table.js

@@ -5,8 +5,7 @@ const Parser = require('..'),
 	Token = require('../src'),
 	TableToken = require('../src/table'),
 	TrToken = require('../src/table/tr'),
-	TdToken = require('../src/table/td'),
-	DdToken = require('../src/nowiki/dd');
+	TdToken = require('../src/table/td');
 
 /**
  * 解析表格，注意`tr`和`td`包含开头的换行
@@ -16,7 +15,9 @@ const Parser = require('..'),
 const parseTable = ({firstChild: {data}, type, name}, config = Parser.getConfig(), accum = []) => {
 	const /** @type {TrToken[]} */ stack = [],
 		lines = data.split('\n');
-	let out = type === 'root' || type === 'ext-inner' && name === 'poem' ? '' : `\n${lines.shift()}`;
+	let out = type === 'root' || type === 'parameter-value' || type === 'ext-inner' && name === 'poem'
+		? ''
+		: `\n${lines.shift()}`;
 
 	/**
 	 * 向表格中插入纯文本
@@ -28,13 +29,13 @@ const parseTable = ({firstChild: {data}, type, name}, config = Parser.getConfig(
 			out += str;
 			return;
 		}
-		const /** @type {Token & {firstChild: AstText}} */ {lastChild} = top;
-		if (lastChild.isPlain()) {
-			lastChild.firstChild.appendData(str);
+		const /** @type {Token}} */ {lastChild} = top;
+		if (lastChild.constructor === Token) {
+			lastChild.setText(String(lastChild) + str);
 		} else {
 			const token = new Token(str, config, true, accum);
 			token.type = 'table-inter';
-			top.appendChild(token.setAttribute('stage', 3));
+			top.insertAt(token.setAttribute('stage', 3));
 		}
 	};
 	for (const outLine of lines) {
@@ -48,6 +49,7 @@ const parseTable = ({firstChild: {data}, type, name}, config = Parser.getConfig(
 			}
 			const [, indent, moreSpaces, tableSyntax, attr] = matchesStart;
 			if (indent) {
+				const DdToken = require('../src/nowiki/dd');
 				new DdToken(indent, config, accum);
 			}
 			push(`\n${spaces}${indent && `\0${accum.length - 1}d\x7F`}${moreSpaces}\0${accum.length}b\x7F`, top);
@@ -82,7 +84,7 @@ const parseTable = ({firstChild: {data}, type, name}, config = Parser.getConfig(
 			}
 			const tr = new TrToken(`\n${spaces}${row}`, attr, config, accum);
 			stack.push(top, tr);
-			top.appendChild(tr);
+			top.insertAt(tr);
 		} else {
 			if (top.type === 'td') {
 				top = stack.pop();
@@ -95,14 +97,14 @@ const parseTable = ({firstChild: {data}, type, name}, config = Parser.getConfig(
 				lastSyntax = `\n${spaces}${cell}`;
 			while (mt) {
 				const td = new TdToken(lastSyntax, attr.slice(lastIndex, mt.index), config, accum);
-				top.appendChild(td);
+				top.insertAt(td);
 				({lastIndex} = regex);
 				[lastSyntax] = mt;
 				mt = regex.exec(attr);
 			}
 			const td = new TdToken(lastSyntax, attr.slice(lastIndex), config, accum);
 			stack.push(top, td);
-			top.appendChild(td);
+			top.insertAt(td);
 		}
 	}
 	return out.slice(1);

package/src/arg.js

@@ -7,11 +7,16 @@ const {text, noWrap} = require('../util/string'),
 
 /**
  * `{{{}}}`包裹的参数
- * @classdesc `{childNodes: [AtomToken, Token, ...HiddenToken]}`
+ * @classdesc `{childNodes: [AtomToken, ?Token, ...HiddenToken]}`
  */
 class ArgToken extends Token {
 	type = 'arg';
 
+	/** default */
+	get default() {
+		return this.childNodes[1]?.text() ?? false;
+	}
+
 	/**
 	 * @param {string[]} parts 以'|'分隔的各部分
 	 * @param {accum} accum
@@ -25,39 +30,16 @@ class ArgToken extends Token {
 				const token = new AtomToken(parts[i], i === 0 ? 'arg-name' : undefined, config, accum, {
 					'Stage-2': ':', '!HeadingToken': '',
 				});
-				this.appendChild(token);
+				super.insertAt(token);
 			} else {
 				const token = new Token(parts[i], config, true, accum);
 				token.type = 'arg-default';
-				this.appendChild(token.setAttribute('stage', 2));
+				super.insertAt(token.setAttribute('stage', 2));
 			}
 		}
 		this.getAttribute('protectChildren')(0);
 	}
 
-	/** @override */
-	cloneNode() {
-		const [name, ...cloned] = this.cloneChildNodes();
-		return Parser.run(() => {
-			const token = new ArgToken([''], this.getAttribute('config'));
-			token.firstChild.safeReplaceWith(name);
-			token.append(...cloned);
-			return token.afterBuild();
-		});
-	}
-
-	/** @override */
-	afterBuild() {
-		this.setAttribute('name', this.firstChild.text().trim());
-		const /** @type {AstListener} */ argListener = ({prevTarget}) => {
-			if (prevTarget === this.firstChild) {
-				this.setAttribute('name', prevTarget.text().trim());
-			}
-		};
-		this.addEventListener(['remove', 'insert', 'replace', 'text'], argListener);
-		return this;
-	}
-
 	/**
 	 * @override
 	 * @param {string} selector
@@ -81,6 +63,47 @@ class ArgToken extends Token {
 		return super.print({pre: '{{{', post: '}}}', sep: '|'});
 	}
 
+	/**
+	 * @override
+	 * @param {number} start 起始位置
+	 * @returns {LintError[]}
+	 */
+	lint(start = 0) {
+		const {childNodes: [argName, argDefault, ...rest]} = this,
+			errors = argName.lint(start + 3);
+		if (argDefault) {
+			errors.push(...argDefault.lint(start + 4 + String(argName).length));
+		}
+		if (rest.length > 0) {
+			const rect = this.getRootNode().posFromIndex(start);
+			errors.push(...rest.map(child => generateForChild(child, rect, '三重括号内的不可见部分')));
+		}
+		return errors;
+	}
+
+	/** @override */
+	cloneNode() {
+		const [name, ...cloned] = this.cloneChildNodes();
+		return Parser.run(() => {
+			const token = new ArgToken([''], this.getAttribute('config'));
+			token.firstChild.safeReplaceWith(name);
+			token.append(...cloned);
+			return token.afterBuild();
+		});
+	}
+
+	/** @override */
+	afterBuild() {
+		this.setAttribute('name', this.firstChild.text().trim());
+		const /** @type {AstListener} */ argListener = ({prevTarget}) => {
+			if (prevTarget === this.firstChild) {
+				this.setAttribute('name', prevTarget.text().trim());
+			}
+		};
+		this.addEventListener(['remove', 'insert', 'replace', 'text'], argListener);
+		return this;
+	}
+
 	/**
 	 * @override
 	 * @returns {string}
@@ -101,24 +124,6 @@ class ArgToken extends Token {
 		});
 	}
 
-	/**
-	 * @override
-	 * @param {number} start 起始位置
-	 * @returns {LintError[]}
-	 */
-	lint(start = 0) {
-		const {childNodes: [argName, argDefault, ...rest]} = this,
-			errors = argName.lint(start + 3);
-		if (argDefault) {
-			errors.push(...argDefault.lint(start + 4 + String(argName).length));
-		}
-		if (rest.length > 0) {
-			const rect = this.getRootNode().posFromIndex(start);
-			errors.push(...rest.map(child => generateForChild(child, rect, '三重括号内的不可见部分')));
-		}
-		return errors;
-	}
-
 	/**
 	 * 移除子节点，且在移除`arg-default`子节点时自动移除全部多余子节点
 	 * @param {number} i 移除位置
@@ -139,7 +144,7 @@ class ArgToken extends Token {
 	 */
 	insertAt(token, i = this.childNodes.length) {
 		const j = i < 0 ? i + this.childNodes.length : i;
-		if (j > 1 && !Parser.running) {
+		if (j > 1) {
 			throw new RangeError(`${this.constructor.name} 不可插入多余的子节点！`);
 		}
 		super.insertAt(token, i);
@@ -157,7 +162,7 @@ class ArgToken extends Token {
 	setName(name) {
 		name = String(name);
 		const root = Parser.parse(`{{{${name}}}}`, this.getAttribute('include'), 2, this.getAttribute('config')),
-			{childNodes: {length}, firstChild: arg} = root;
+			{length, firstChild: arg} = root;
 		if (length !== 1 || arg.type !== 'arg' || arg.childNodes.length !== 1) {
 			throw new SyntaxError(`非法的参数名称：${noWrap(name)}`);
 		}
@@ -174,7 +179,7 @@ class ArgToken extends Token {
 	setDefault(value) {
 		value = String(value);
 		const root = Parser.parse(`{{{|${value}}}}`, this.getAttribute('include'), 2, this.getAttribute('config')),
-			{childNodes: {length}, firstChild: arg} = root;
+			{length, firstChild: arg} = root;
 		if (length !== 1 || arg.type !== 'arg' || arg.childNodes.length !== 2) {
 			throw new SyntaxError(`非法的参数预设值：${noWrap(value)}`);
 		}
@@ -184,7 +189,7 @@ class ArgToken extends Token {
 		if (oldDefault) {
 			oldDefault.safeReplaceWith(lastChild);
 		} else {
-			this.appendChild(lastChild);
+			this.insertAt(lastChild);
 		}
 	}
 }

package/src/atom/index.js

@@ -5,7 +5,7 @@ const Parser = require('../..'),
 
 /**
  * 不会被继续解析的plain Token
- * @classdesc `{childNodes: (AstText|Token)[]}`
+ * @classdesc `{childNodes: ...AstText|Token}`
  */
 class AtomToken extends Token {
 	type = 'plain';
@@ -30,10 +30,12 @@ class AtomToken extends Token {
 	cloneNode() {
 		const cloned = this.cloneChildNodes(),
 			config = this.getAttribute('config'),
-			acceptable = this.getAttribute('acceptable'),
-			token = Parser.run(() => new this.constructor(undefined, this.type, config, [], acceptable));
-		token.append(...cloned);
-		return token;
+			acceptable = this.getAttribute('acceptable');
+		return Parser.run(() => {
+			const token = new this.constructor(undefined, this.type, config, [], acceptable);
+			token.append(...cloned);
+			return token;
+		});
 	}
 }