wikiparser-node 0.4.0 → 0.6.2-b

package/parser/hrAndDoubleUnderscore.js

@@ -1,35 +0,0 @@
-'use strict';
-
-const Parser = require('..'),
-	AstText = require('../lib/text');
-
-/**
- * 解析\<hr\>和状态开关
- * @param {{firstChild: AstText, type: string}} root 根节点
- * @param {accum} accum
- */
-const parseHrAndDoubleUnderscore = ({firstChild: {data}, type}, config = Parser.getConfig(), accum = []) => {
-	const HrToken = require('../src/nowiki/hr'),
-		DoubleUnderscoreToken = require('../src/nowiki/doubleUnderscore');
-	const {doubleUnderscore} = config;
-	if (type !== 'root') {
-		data = `\0${data}`;
-	}
-	data = data.replaceAll(/^((?:\0\d+c\x7F)*)(-{4,})/gmu, (_, lead, m) => {
-		new HrToken(m.length, config, accum);
-		return `${lead}\0${accum.length - 1}r\x7F`;
-	}).replaceAll(
-		new RegExp(`__(${doubleUnderscore.flat().join('|')})__`, 'giu'),
-		/** @param {string} p1 */ (m, p1) => {
-			if (doubleUnderscore[0].includes(p1.toLowerCase()) || doubleUnderscore[1].includes(p1)) {
-				new DoubleUnderscoreToken(p1, config, accum);
-				return `\0${accum.length - 1}u\x7F`;
-			}
-			return m;
-		},
-	);
-	return type === 'root' ? data : data.slice(1);
-};
-
-Parser.parsers.parseHrAndDoubleUnderscore = __filename;
-module.exports = parseHrAndDoubleUnderscore;

package/parser/html.js

@@ -1,42 +0,0 @@
-'use strict';
-
-const Parser = require('..');
-
-/**
- * 解析HTML标签
- * @param {string} wikitext wikitext
- * @param {accum} accum
- */
-const parseHtml = (wikitext, config = Parser.getConfig(), accum = []) => {
-	const regex = /^(\/?)([a-z][^\s/>]*)(\s[^>]*?)?(\/?>)([^<]*)$/iu,
-		elements = config.html.flat(),
-		bits = wikitext.split('<');
-	let text = bits.shift();
-	for (const x of bits) {
-		const mt = regex.exec(x),
-			t = mt?.[2],
-			name = t?.toLowerCase();
-		if (!mt || !elements.includes(name)) {
-			text += `<${x}`;
-			continue;
-		}
-		const AttributeToken = require('../src/attribute');
-		const [, slash,, params = '', brace, rest] = mt,
-			attr = new AttributeToken(params, 'html-attr', name, config, accum),
-			itemprop = attr.getAttr('itemprop');
-		if (name === 'meta' && (itemprop === undefined || attr.getAttr('content') === undefined)
-			|| name === 'link' && (itemprop === undefined || attr.getAttr('href') === undefined)
-		) {
-			text += `<${x}`;
-			accum.pop();
-			continue;
-		}
-		text += `\0${accum.length}x\x7F${rest}`;
-		const HtmlToken = require('../src/html');
-		new HtmlToken(t, attr, slash === '/', brace === '/>', config, accum);
-	}
-	return text;
-};
-
-Parser.parsers.parseHtml = __filename;
-module.exports = parseHtml;

package/parser/links.js

@@ -1,98 +0,0 @@
-'use strict';
-
-const Parser = require('..'),
-	Token = require('../src');
-
-/**
- * 解析内部链接
- * @param {string} wikitext wikitext
- * @param {accum} accum
- */
-const parseLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
-	const parseQuotes = require('./quotes.js');
-	const regex = /^([^\n<>[\]{}|]+)(?:\|(.*?[^\]]))?\]\](.*)$/su,
-		regexImg = /^([^\n<>[\]{}|]+)\|(.*)$/su,
-		regexExt = new RegExp(`^\\s*(?:${config.protocol})`, 'iu'),
-		bits = wikitext.split('[[');
-	let s = bits.shift();
-	for (let i = 0; i < bits.length; i++) {
-		let mightBeImg, link, text, after;
-		const x = bits[i],
-			m = regex.exec(x);
-		if (m) {
-			[, link, text, after] = m;
-			if (after[0] === ']' && text?.includes('[')) {
-				text += ']';
-				after = after.slice(1);
-			}
-		} else {
-			const m2 = regexImg.exec(x);
-			if (m2) {
-				mightBeImg = true;
-				[, link, text] = m2;
-			}
-		}
-		if (link === undefined || regexExt.test(link) || /\0\d+[exhbru]\x7F/u.test(link)) {
-			s += `[[${x}`;
-			continue;
-		}
-		let page = link;
-		if (link.includes('%')) {
-			try {
-				page = decodeURIComponent(link);
-			} catch {}
-		}
-		const force = link.trim()[0] === ':';
-		if (force && mightBeImg) {
-			s += `[[${x}`;
-			continue;
-		}
-		const title = Parser.normalizeTitle(page, 0, false, config, true),
-			{ns, interwiki, valid} = title;
-		if (!valid) {
-			s += `[[${x}`;
-			continue;
-		} else if (mightBeImg) {
-			if (interwiki || ns !== 6) {
-				s += `[[${x}`;
-				continue;
-			}
-			let found;
-			for (i++; i < bits.length; i++) {
-				const next = bits[i],
-					p = next.split(']]');
-				if (p.length > 2) {
-					found = true;
-					text += `[[${p[0]}]]${p[1]}`;
-					after = p.slice(2).join(']]');
-					break;
-				} else if (p.length === 2) {
-					text += `[[${p[0]}]]${p[1]}`;
-				} else {
-					text += `[[${next}`;
-					break;
-				}
-			}
-			text = parseLinks(text, config, accum);
-			if (!found) {
-				s += `[[${link}|${text}`;
-				continue;
-			}
-		}
-		text &&= parseQuotes(text, config, accum);
-		s += `\0${accum.length}l\x7F${after}`;
-		let LinkToken = require('../src/link');
-		if (!force) {
-			if (!interwiki && ns === 6) {
-				LinkToken = require('../src/link/file');
-			} else if (!interwiki && ns === 14) {
-				LinkToken = require('../src/link/category');
-			}
-		}
-		new LinkToken(link, text, title, config, accum);
-	}
-	return s;
-};
-
-Parser.parsers.parseLinks = __filename;
-module.exports = parseLinks;

package/parser/list.js

@@ -1,59 +0,0 @@
-'use strict';
-
-const Parser = require('..');
-
-/**
- * 解析列表
- * @param {string} text wikitext
- * @param {accum} accum
- */
-const parseList = (text, config = Parser.getConfig(), accum = []) => {
-	const mt = /^((?:\0\d+c\x7F)*)([;:*#]+)/u.exec(text);
-	if (!mt) {
-		return text;
-	}
-	const ListToken = require('../src/nowiki/list');
-	const [total, comment, prefix] = mt;
-	text = `${comment}\0${accum.length}d\x7F${text.slice(total.length)}`;
-	new ListToken(prefix, config, accum);
-	let dt = prefix.split(';').length - 1;
-	if (!dt) {
-		return text;
-	}
-	const DdToken = require('../src/nowiki/dd');
-	let regex = /:+|-\{/gu,
-		ex = regex.exec(text),
-		lc = 0;
-	while (ex && dt) {
-		const {0: syntax, index} = ex;
-		if (syntax[0] === ':') {
-			if (syntax.length >= dt) {
-				new DdToken(':'.repeat(dt), config, accum);
-				return `${text.slice(0, index)}\0${accum.length - 1}d\x7F${text.slice(index + dt)}`;
-			}
-			text = `${text.slice(0, index)}\0${accum.length}d\x7F${text.slice(regex.lastIndex)}`;
-			dt -= syntax.length;
-			regex.lastIndex = index + 4 + String(accum.length).length;
-			new DdToken(syntax, config, accum);
-		} else if (syntax === '-{') {
-			if (!lc) {
-				const {lastIndex} = regex;
-				regex = /-\{|\}-/gu;
-				regex.lastIndex = lastIndex;
-			}
-			lc++;
-		} else {
-			lc--;
-			if (!lc) {
-				const {lastIndex} = regex;
-				regex = /:+|-\{/gu;
-				regex.lastIndex = lastIndex;
-			}
-		}
-		ex = regex.exec(text);
-	}
-	return text;
-};
-
-Parser.parsers.parseList = __filename;
-module.exports = parseList;

package/parser/magicLinks.js

@@ -1,41 +0,0 @@
-'use strict';
-
-const {extUrlChar} = require('../util/string'),
-	Parser = require('..');
-
-/**
- * 解析自由外链
- * @param {string} wikitext wikitext
- * @param {accum} accum
- */
-const parseMagicLinks = (wikitext, config = Parser.getConfig(), accum = []) => {
-	const MagicLinkToken = require('../src/magicLink');
-	const regex = new RegExp(`\\b(?:${config.protocol})(${extUrlChar})`, 'giu');
-	return wikitext.replace(regex, /** @param {string} p1 */ (m, p1) => {
-		let trail = '',
-			url = m;
-		const m2 = /&(?:lt|gt|nbsp|#x0*(?:3[ce]|a0)|#0*(?:6[02]|160));/iu.exec(url);
-		if (m2) {
-			trail = url.slice(m2.index);
-			url = url.slice(0, m2.index);
-		}
-		const sep = new RegExp(`[,;.:!?${url.includes('(') ? '' : ')'}]+$`, 'u'),
-			sepChars = sep.exec(url);
-		if (sepChars) {
-			let correction = 0;
-			if (sepChars[0][0] === ';' && /&(?:[a-z]+|#x[\da-f]+|#\d+)$/iu.test(url.slice(0, sepChars.index))) {
-				correction = 1;
-			}
-			trail = `${url.slice(sepChars.index + correction)}${trail}`;
-			url = url.slice(0, sepChars.index + correction);
-		}
-		if (trail.length >= p1.length) {
-			return m;
-		}
-		new MagicLinkToken(url, false, config, accum);
-		return `\0${accum.length - 1}w\x7F${trail}`;
-	});
-};
-
-Parser.parsers.parseMagicLinks = __filename;
-module.exports = parseMagicLinks;

package/parser/quotes.js

@@ -1,64 +0,0 @@
-'use strict';
-
-const Parser = require('..');
-
-/**
- * 解析单引号
- * @param {string} text wikitext
- * @param {accum} accum
- */
-const parseQuotes = (text, config = Parser.getConfig(), accum = []) => {
-	const arr = text.split(/('{2,})/u),
-		{length} = arr;
-	if (length === 1) {
-		return text;
-	}
-	let nBold = 0,
-		nItalic = 0,
-		firstSingle, firstMulti, firstSpace;
-	for (let i = 1; i < length; i += 2) {
-		const {length: len} = arr[i];
-		switch (len) {
-			case 2:
-				nItalic++;
-				break;
-			case 4:
-				arr[i - 1] += "'";
-				arr[i] = "'''";
-				// fall through
-			case 3:
-				nBold++;
-				if (firstSingle) {
-					break;
-				} else if (arr[i - 1].at(-1) === ' ') {
-					if (!firstMulti && !firstSpace) {
-						firstSpace = i;
-					}
-				} else if (arr[i - 1].at(-2) === ' ') {
-					firstSingle = i;
-				} else {
-					firstMulti ||= i;
-				}
-				break;
-			default:
-				arr[i - 1] += "'".repeat(len - 5);
-				arr[i] = "'''''";
-				nItalic++;
-				nBold++;
-		}
-	}
-	if (nItalic % 2 === 1 && nBold % 2 === 1) {
-		const i = firstSingle ?? firstMulti ?? firstSpace;
-		arr[i] = "''";
-		arr[i - 1] += "'";
-	}
-	const QuoteToken = require('../src/nowiki/quote');
-	for (let i = 1; i < length; i += 2) {
-		new QuoteToken(arr[i].length, config, accum);
-		arr[i] = `\0${accum.length - 1}q\x7F`;
-	}
-	return arr.join('');
-};
-
-Parser.parsers.parseQuotes = __filename;
-module.exports = parseQuotes;

package/parser/selector.js

@@ -1,175 +0,0 @@
-'use strict';
-
-const Parser = require('..');
-
-const /** @type {pseudo[]} */ simplePseudos = [
-		'root',
-		'first-child',
-		'first-of-type',
-		'last-child',
-		'last-of-type',
-		'only-child',
-		'only-of-type',
-		'empty',
-		'parent',
-		'header',
-		'hidden',
-		'visible',
-		'only-whitespace',
-		'local-link',
-		'read-only',
-		'read-write',
-		'invalid',
-		'required',
-		'optional',
-	],
-	/** @type {pseudo[]} */ complexPseudos = [
-		'is',
-		'not',
-		'nth-child',
-		'nth-of-type',
-		'nth-last-child',
-		'nth-last-of-type',
-		'contains',
-		'has',
-		'lang',
-	],
-	specialChars = [
-		['[', '['],
-		[']', ']'],
-		['(', '('],
-		[')', ')'],
-		['"', '"'],
-		["'", '''],
-		[':', ':'],
-		['\\', '\'],
-		['&', '&'],
-	],
-	pseudoRegex = new RegExp(`:(${complexPseudos.join('|')})$`, 'u'),
-	regularRegex = /[[(,>+~]|\s+/u, // eslint-disable-line regexp/no-super-linear-move
-	attributeRegex = /^\s*(\w+)\s*(?:([~|^$*!]?=)\s*("[^"]*"|'[^']*'|[^\s[\]]+)(?:\s+(i))?\s*)?\]/u,
-	functionRegex = /^(\s*"[^"]*"\s*|\s*'[^']*'\s*|[^()]*)\)/u;
-
-/**
- * 清理转义符号
- * @param {string} selector
- */
-const sanitize = selector => {
-	for (const [c, escaped] of specialChars) {
-		selector = selector.replaceAll(`\\${c}`, escaped);
-	}
-	return selector;
-};
-
-/**
- * 还原转义符号
- * @param {string|undefined} selector
- */
-const desanitize = selector => {
-	if (selector === undefined) {
-		return undefined;
-	}
-	for (const [c, escaped] of specialChars) {
-		selector = selector.replaceAll(escaped, c);
-	}
-	return selector.trim();
-};
-
-/**
- * 去除首尾的引号
- * @param {string|undefined} val 属性值或伪选择器函数的参数
- */
-const deQuote = val => {
-	if (val === undefined) {
-		return undefined;
-	}
-	const quotes = /^(["']).*\1$/u.exec(val)?.[1];
-	return quotes ? val.slice(1, -1) : val;
-};
-
-/**
- * 解析简单伪选择器
- * @param {SelectorArray} step 当前顶部
- * @param {string} str 不含属性和复杂伪选择器的语句
- * @throws `SyntaxError` 非法的选择器
- */
-const pushSimple = (step, str) => {
-	const pieces = str.trim().split(':'),
-		// eslint-disable-next-line unicorn/explicit-length-check
-		i = pieces.slice(1).findIndex(pseudo => simplePseudos.includes(pseudo)) + 1 || pieces.length;
-	if (pieces.slice(i).some(pseudo => !simplePseudos.includes(pseudo))) {
-		throw new SyntaxError(`非法的选择器！\n${str}\n可能需要将':'转义为'\\:'。`);
-	}
-	step.push(desanitize(pieces.slice(0, i).join(':')), ...pieces.slice(i).map(piece => `:${piece}`));
-};
-
-/**
- * 解析选择器
- * @param {string} selector
- * @throws `SyntaxError` 非法的选择器
- */
-const parseSelector = selector => {
-	selector = selector.trim();
-	const /** @type {SelectorArray[][]} */ stack = [[[]]];
-	let sanitized = sanitize(selector),
-		regex = regularRegex,
-		mt = regex.exec(sanitized),
-		[condition] = stack,
-		[step] = condition;
-	while (mt) {
-		let {0: syntax, index} = mt;
-		if (syntax.trim() === '') {
-			index += syntax.length;
-			const char = sanitized[index];
-			syntax = [',', '>', '+', '~'].includes(char) ? char : '';
-		}
-		if (syntax === ',') { // 情形1：并列
-			pushSimple(step, sanitized.slice(0, index));
-			condition = [[]];
-			[step] = condition;
-			stack.push(condition);
-		} else if (['>', '+', '~', ''].includes(syntax)) { // 情形2：关系
-			pushSimple(step, sanitized.slice(0, index));
-			if (!step.some(Boolean)) {
-				throw new SyntaxError(`非法的选择器！\n${selector}\n可能需要通用选择器'*'。`);
-			}
-			step.relation = syntax;
-			step = [];
-			condition.push(step);
-		} else if (syntax === '[') { // 情形3：属性开启
-			pushSimple(step, sanitized.slice(0, index));
-			regex = attributeRegex;
-		} else if (syntax.at(-1) === ']') { // 情形4：属性闭合
-			mt[3] = desanitize(deQuote(mt[3]));
-			step.push(mt.slice(1));
-			regex = regularRegex;
-		} else if (syntax === '(') { // 情形5：伪选择器开启
-			const pseudoExec = pseudoRegex.exec(sanitized.slice(0, index));
-			if (!pseudoExec) {
-				throw new SyntaxError(`非法的选择器！\n${desanitize(sanitized)}\n请检查伪选择器是否存在。`);
-			}
-			pushSimple(step, sanitized.slice(0, pseudoExec.index));
-			step.push(pseudoExec[1]); // 临时存放复杂伪选择器
-			regex = functionRegex;
-		} else { // 情形6：伪选择器闭合
-			const /** @type {pseudo} */ pseudo = step.pop();
-			mt.push(pseudo);
-			mt[1] = deQuote(mt[1]);
-			step.push(mt.slice(1));
-			regex = regularRegex;
-		}
-		sanitized = sanitized.slice(index + syntax.length);
-		if ([',', '>', '+', '~'].includes(syntax)) {
-			sanitized = sanitized.trim();
-		}
-		mt = regex.exec(sanitized);
-	}
-	if (regex !== regularRegex) {
-		throw new SyntaxError(`非法的选择器！\n${selector}\n检测到未闭合的'${regex === attributeRegex ? '[' : '('}'`);
-	}
-	pushSimple(step, sanitized);
-	return stack;
-};
-
-Parser.parsers.parseSelector = __filename;
-module.exports = parseSelector;

package/parser/table.js

@@ -1,112 +0,0 @@
-'use strict';
-
-const Parser = require('..'),
-	AstText = require('../lib/text');
-
-/**
- * 解析表格，注意`tr`和`td`包含开头的换行
- * @param {{firstChild: AstText, type: string}} root 根节点
- * @param {accum} accum
- */
-const parseTable = ({firstChild: {data}, type}, config = Parser.getConfig(), accum = []) => {
-	const Token = require('../src'),
-		TableToken = require('../src/table'),
-		TrToken = require('../src/table/tr'),
-		TdToken = require('../src/table/td'),
-		DdToken = require('../src/nowiki/dd');
-	const /** @type {TrToken[]} */ stack = [],
-		lines = data.split('\n');
-	let out = type === 'root' ? '' : `\n${lines.shift()}`;
-
-	/**
-	 * 向表格中插入纯文本
-	 * @param {string} str 待插入的文本
-	 * @param {TrToken} top 当前解析的表格或表格行
-	 */
-	const push = (str, top) => {
-		if (!top) {
-			out += str;
-			return;
-		}
-		const {lastElementChild} = top;
-		if (lastElementChild.isPlain()) {
-			lastElementChild.setText(String(lastElementChild) + str);
-		} else {
-			const token = new Token(str, config, true, accum);
-			token.type = 'table-inter';
-			top.appendChild(token.setAttribute('stage', 3));
-		}
-	};
-	for (const outLine of lines) {
-		let top = stack.pop();
-		const [spaces] = /^(?:\s|\0\d+c\x7F)*/u.exec(outLine),
-			line = outLine.slice(spaces.length),
-			matchesStart = /^(:*)((?:\s|\0\d+c\x7F)*)(\{\||\{\0\d+!\x7F|\0\d+\{\x7F)(.*)$/u.exec(line);
-		if (matchesStart) {
-			while (top && top.type !== 'td') {
-				top = stack.pop();
-			}
-			const [, indent, moreSpaces, tableSyntax, attr] = matchesStart;
-			if (indent) {
-				new DdToken(indent, config, accum);
-			}
-			push(`\n${spaces}${indent && `\0${accum.length - 1}d\x7F`}${moreSpaces}\0${accum.length}b\x7F`, top);
-			const table = new TableToken(tableSyntax, attr, config, accum);
-			stack.push(...top ? [top] : [], table);
-			continue;
-		} else if (!top) {
-			out += `\n${outLine}`;
-			continue;
-		}
-		const matches = // eslint-disable-line operator-linebreak
-			/^(?:(\|\}|\0\d+!\x7F\}|\0\d+\}\x7F)|(\|-+|\0\d+!\x7F-+|\0\d+-\x7F-*)(?!-)|(!|(?:\||\0\d+!\x7F)\+?))(.*)$/u
-				.exec(line);
-		if (!matches) {
-			push(`\n${outLine}`, top);
-			stack.push(...top ? [top] : []);
-			continue;
-		}
-		const [, closing, row, cell, attr] = matches;
-		if (closing) {
-			while (!(top instanceof TableToken)) {
-				top = stack.pop();
-			}
-			top.close(`\n${spaces}${closing}`, true);
-			push(attr, stack.at(-1));
-		} else if (row) {
-			if (top.type === 'td') {
-				top = stack.pop();
-			}
-			if (top.type === 'tr') {
-				top = stack.pop();
-			}
-			const tr = new TrToken(`\n${spaces}${row}`, attr, config, accum);
-			stack.push(top, tr);
-			top.appendChild(tr);
-		} else {
-			if (top.type === 'td') {
-				top = stack.pop();
-			}
-			const regex = cell === '!'
-				? /!!|(?:\||\0\d+!\x7F){2}|\0\d+\+\x7F/gu
-				: /(?:\||\0\d+!\x7F){2}|\0\d+\+\x7F/gu;
-			let mt = regex.exec(attr),
-				lastIndex = 0,
-				lastSyntax = `\n${spaces}${cell}`;
-			while (mt) {
-				const td = new TdToken(lastSyntax, attr.slice(lastIndex, mt.index), config, accum);
-				top.appendChild(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);
-		}
-	}
-	return out.slice(1);
-};
-
-Parser.parsers.parseTable = __filename;
-module.exports = parseTable;

package/printed/README

@@ -1 +0,0 @@
-这里存放以 JSON 格式打印的 AST。