wikiparser-node 0.9.2-b → 0.10.0-m

package/src/index.js

@@ -0,0 +1,406 @@
+'use strict';
+
+// PHP解析器的步骤：
+// -1. 替换签名和`{{subst:}}`，参见Parser::preSaveTransform；这在revision中不可能保留，可以跳过
+// 0. 移除特定字符`\0`和`\x7F`，参见Parser::parse
+// 1. 注释/扩展标签（'<'相关），参见Preprocessor_Hash::buildDomTreeArrayFromText和Sanitizer::decodeTagAttributes
+// 2. 模板/模板变量/标题，注意rightmost法则，以及`-{`和`[[`可以破坏`{{`或`{{{`语法，
+//    参见Preprocessor_Hash::buildDomTreeArrayFromText
+// 3. HTML标签（允许不匹配），参见Sanitizer::internalRemoveHtmlTags
+// 4. 表格，参见Parser::handleTables
+// 5. 水平线、状态开关和余下的标题，参见Parser::internalParse
+// 6. 内链，含文件和分类，参见Parser::handleInternalLinks2
+// 7. `'`，参见Parser::doQuotes
+// 8. 外链，参见Parser::handleExternalLinks
+// 9. ISBN、RFC（未来将废弃，不予支持）和自由外链，参见Parser::handleMagicLinks
+// 10. 段落和列表，参见BlockLevelPass::execute
+// 11. 转换，参见LanguageConverter::recursiveConvertTopLevel
+
+// \0\d+.\x7F标记Token：
+// e: ExtToken
+// a: AttributeToken
+// c: CommentToken、NoIncludeToken和IncludeToken
+// !: `{{!}}`专用
+// {: `{{(!}}`专用
+// }: `{{!)}}`专用
+// -: `{{!-}}`专用
+// +: `{{!!}}`专用
+// ~: `{{=}}`专用
+// s: `{{{|subst:}}}`
+// m: `{{fullurl:}}`、`{{canonicalurl:}}`或`{{filepath:}}`
+// t: ArgToken或TranscludeToken
+// h: HeadingToken
+// x: HtmlToken
+// b: TableToken
+// r: HrToken
+// u: DoubleUnderscoreToken
+// l: LinkToken
+// q: QuoteToken
+// w: ExtLinkToken
+// d: ListToken
+// v: ConverterToken
+
+const {text} = require('../util/string'),
+	Parser = require('..'),
+	AstElement = require('../lib/element'),
+	AstText = require('../lib/text');
+const {MAX_STAGE} = Parser;
+
+/**
+ * 所有节点的基类
+ * @classdesc `{childNodes: ...(AstText|Token)}`
+ */
+class Token extends AstElement {
+	type = 'root';
+	#stage = 0; // 解析阶段，参见顶部注释。只对plain Token有意义。
+	#config;
+	// 这个数组起两个作用：1. 数组中的Token会在build时替换`/\0\d+.\x7F/`标记；2. 数组中的Token会依次执行parseOnce和build方法。
+	#accum;
+	/** @type {boolean} */ #include;
+
+	/**
+	 * 将维基语法替换为占位符
+	 * @param {number} n 解析阶段
+	 * @param {boolean} include 是否嵌入
+	 */
+	#parseOnce = (n = this.#stage, include = false) => {
+		if (n < this.#stage || !this.isPlain() || this.length === 0) {
+			return this;
+		}
+		switch (n) {
+			case 0:
+				if (this.type === 'root') {
+					this.#accum.shift();
+				}
+				this.#include = Boolean(include);
+				this.#parseCommentAndExt(include);
+				break;
+			case 1:
+				this.#parseBrackets();
+				break;
+			case 2:
+				this.#parseHtml();
+				break;
+			case 3:
+				this.#parseTable();
+				break;
+			case 4:
+				this.#parseHrAndDoubleUndescore();
+				break;
+			case 5:
+				this.#parseLinks();
+				break;
+			case 6:
+				this.#parseQuotes();
+				break;
+
+			case 7:
+				this.#parseExternalLinks();
+				break;
+			case 8:
+				this.#parseMagicLinks();
+				break;
+			case 9:
+				this.#parseList();
+				break;
+			case 10:
+				this.#parseConverter();
+				// no default
+		}
+		if (this.type === 'root') {
+			for (const token of this.#accum) {
+				token.getAttribute('parseOnce')(n, include);
+			}
+		}
+		this.#stage++;
+		return this;
+	};
+
+	/**
+	 * 重建wikitext
+	 * @template {string} T
+	 * @param {string} str 半解析的字符串
+	 * @param {T} type 返回类型
+	 * @complexity `n`
+	 * @returns {T extends 'string|text' ? string : (Token|AstText)[]}
+	 */
+	#buildFromStr = (str, type) => {
+		const nodes = str.split(/[\0\x7F]/u).map((s, i) => {
+			if (i % 2 === 0) {
+				return new AstText(s);
+			} else if (isNaN(s.at(-1))) {
+				return this.#accum[Number(s.slice(0, -1))];
+			}
+			throw new Error(`解析错误！未正确标记的 Token：${s}`);
+		});
+		if (type === 'string') {
+			return nodes.map(String).join('');
+		} else if (type === 'text') {
+			return text(nodes);
+		}
+		return nodes;
+	};
+
+	/**
+	 * 将占位符替换为子Token
+	 * @complexity `n`
+	 */
+	#build = () => {
+		this.#stage = MAX_STAGE;
+		const {length, firstChild} = this,
+			str = String(firstChild);
+		if (length === 1 && firstChild.type === 'text' && str.includes('\0')) {
+			this.replaceChildren(...this.#buildFromStr(str));
+			this.normalize();
+			if (this.type === 'root') {
+				for (const token of this.#accum) {
+					token.getAttribute('build')();
+				}
+			}
+		}
+	};
+
+	/**
+	 * @param {string} wikitext wikitext
+	 * @param {accum} accum
+	 */
+	constructor(wikitext, config = Parser.getConfig(), halfParsed = false, accum = [], acceptable = undefined) {
+		super();
+		if (typeof wikitext === 'string') {
+			this.insertAt(halfParsed ? wikitext : wikitext.replace(/[\0\x7F]/gu, ''));
+		}
+		this.#config = config;
+		this.#accum = accum;
+		accum.push(this);
+	}
+
+	/**
+	 * @override
+	 * @template {string} T
+	 * @param {T} key 属性键
+	 * @returns {TokenAttribute<T>}
+	 */
+	getAttribute(key) {
+		switch (key) {
+			case 'config':
+				return structuredClone(this.#config);
+			case 'accum':
+				return this.#accum;
+			case 'parseOnce':
+				return this.#parseOnce;
+			case 'buildFromStr':
+				return this.#buildFromStr;
+			case 'build':
+				return this.#build;
+			case 'include': {
+				if (this.#include !== undefined) {
+					return this.#include;
+				}
+				const root = this.getRootNode();
+				if (root.type === 'root' && root !== this) {
+					return root.getAttribute('include');
+				}
+				return false;
+			}
+			default:
+				return super.getAttribute(key);
+		}
+	}
+
+	/**
+	 * @override
+	 * @template {string} T
+	 * @param {T} key 属性键
+	 * @param {TokenAttribute<T>} value 属性值
+	 */
+	setAttribute(key, value) {
+		switch (key) {
+			case 'stage':
+				if (this.#stage === 0 && this.type === 'root') {
+					this.#accum.shift();
+				}
+				this.#stage = value;
+				return this;
+			default:
+				return super.setAttribute(key, value);
+		}
+	}
+
+	/** 是否是普通节点 */
+	isPlain() {
+		return this.constructor === Token;
+	}
+
+	/**
+	 * @override
+	 * @template {string|Token} T
+	 * @param {T} token 待插入的子节点
+	 * @param {number} i 插入位置
+	 * @complexity `n`
+	 * @returns {T extends Token ? Token : AstText}
+	 */
+	insertAt(token, i = this.length) {
+		if (typeof token === 'string') {
+			token = new AstText(token);
+		}
+		super.insertAt(token, i);
+		if (token.type === 'root') {
+			token.type = 'plain';
+		}
+		return token;
+	}
+
+	/**
+	 * 规范化页面标题
+	 * @param {string} title 标题（含或不含命名空间前缀）
+	 * @param {number} defaultNs 命名空间
+	 * @param {boolean} decode 是否需要解码
+	 * @param {boolean} selfLink 是否允许selfLink
+	 */
+	normalizeTitle(title, defaultNs = 0, halfParsed = false, decode = false, selfLink = false) {
+		return Parser.normalizeTitle(title, defaultNs, this.#include, this.#config, halfParsed, decode, selfLink);
+	}
+
+	/** 生成部分Token的`name`属性 */
+	afterBuild() {
+		if (this.type === 'root') {
+			for (const token of this.#accum) {
+				token.afterBuild();
+			}
+		}
+	}
+
+	/**
+	 * 解析、重构、生成部分Token的`name`属性
+	 * @param {number} n 最大解析层级
+	 * @param {boolean} include 是否嵌入
+	 */
+	parse(n = MAX_STAGE, include = false) {
+		while (this.#stage < n) {
+			this.#parseOnce(this.#stage, include);
+		}
+		if (n) {
+			this.#build();
+			this.afterBuild();
+		}
+		return this;
+	}
+
+	/**
+	 * 解析HTML注释和扩展标签
+	 * @param {boolean} includeOnly 是否嵌入
+	 */
+	#parseCommentAndExt(includeOnly) {
+		const parseCommentAndExt = require('../parser/commentAndExt');
+		this.setText(parseCommentAndExt(String(this.firstChild), this.#config, this.#accum, includeOnly));
+	}
+
+	/** 解析花括号 */
+	#parseBrackets() {
+		const parseBrackets = require('../parser/brackets');
+		const str = this.type === 'root' ? String(this.firstChild) : `\0${String(this.firstChild)}`,
+			parsed = parseBrackets(str, this.#config, this.#accum);
+		this.setText(this.type === 'root' ? parsed : parsed.slice(1));
+	}
+
+	/** 解析HTML标签 */
+	#parseHtml() {
+		if (this.#config.excludes.includes('html')) {
+			return;
+		}
+		const parseHtml = require('../parser/html');
+		this.setText(parseHtml(String(this.firstChild), this.#config, this.#accum));
+	}
+
+	/** 解析表格 */
+	#parseTable() {
+		if (this.#config.excludes.includes('table')) {
+			return;
+		}
+		const parseTable = require('../parser/table'),
+			TableToken = require('./table');
+		this.setText(parseTable(this, this.#config, this.#accum));
+		for (const table of this.#accum) {
+			if (table instanceof TableToken && table.type !== 'td') {
+				table.normalize();
+				const {childNodes: [, child]} = table;
+				if (typeof child === 'string' && child.includes('\0')) {
+					table.removeAt(1);
+					const inner = new Token(child, this.#config, true, this.#accum);
+					table.insertAt(inner, 1);
+					inner.setAttribute('stage', 4);
+				}
+			}
+		}
+	}
+
+	/** 解析\<hr\>和状态开关 */
+	#parseHrAndDoubleUndescore() {
+		if (this.#config.excludes.includes('hr')) {
+			return;
+		}
+		const parseHrAndDoubleUnderscore = require('../parser/hrAndDoubleUnderscore');
+		this.setText(parseHrAndDoubleUnderscore(this, this.#config, this.#accum));
+	}
+
+	/** 解析内部链接 */
+	#parseLinks() {
+		const parseLinks = require('../parser/links');
+		this.setText(parseLinks(String(this.firstChild), this.#config, this.#accum));
+	}
+
+	/** 解析单引号 */
+	#parseQuotes() {
+		if (this.#config.excludes.includes('quote')) {
+			return;
+		}
+		const parseQuotes = require('../parser/quotes');
+		const lines = String(this.firstChild).split('\n');
+		for (let i = 0; i < lines.length; i++) {
+			lines[i] = parseQuotes(lines[i], this.#config, this.#accum);
+		}
+		this.setText(lines.join('\n'));
+	}
+
+	/** 解析外部链接 */
+	#parseExternalLinks() {
+		if (this.#config.excludes.includes('extLink')) {
+			return;
+		}
+		const parseExternalLinks = require('../parser/externalLinks');
+		this.setText(parseExternalLinks(String(this.firstChild), this.#config, this.#accum));
+	}
+
+	/** 解析自由外链 */
+	#parseMagicLinks() {
+		if (this.#config.excludes.includes('magicLink')) {
+			return;
+		}
+		const parseMagicLinks = require('../parser/magicLinks');
+		this.setText(parseMagicLinks(String(this.firstChild), this.#config, this.#accum));
+	}
+
+	/** 解析列表 */
+	#parseList() {
+		if (this.#config.excludes.includes('list')) {
+			return;
+		}
+		const parseList = require('../parser/list');
+		const lines = String(this.firstChild).split('\n');
+		let i = this.type === 'root' || this.type === 'ext-inner' && this.name === 'poem' ? 0 : 1;
+		for (; i < lines.length; i++) {
+			lines[i] = parseList(lines[i], this.#config, this.#accum);
+		}
+		this.setText(lines.join('\n'));
+	}
+
+	/** 解析语言变体转换 */
+	#parseConverter() {
+		if (this.#config.variants?.length > 0) {
+			const parseConverter = require('../parser/converter');
+			this.setText(parseConverter(String(this.firstChild), this.#config, this.#accum));
+		}
+	}
+}
+
+module.exports = Token;

package/src/link/category.js

@@ -0,0 +1,13 @@
+'use strict';
+
+const LinkToken = require('.');
+
+/**
+ * 分类
+ * @classdesc `{childNodes: [AtomToken, ?Token]}`
+ */
+class CategoryToken extends LinkToken {
+	type = 'category';
+}
+
+module.exports = CategoryToken;

package/src/link/file.js

@@ -0,0 +1,132 @@
+'use strict';
+
+const {explode} = require('../../util/string'),
+	{generateForChild} = require('../../util/lint'),
+	Parser = require('../..'),
+	LinkToken = require('.'),
+	ImageParameterToken = require('../imageParameter');
+
+const frame = new Set(['manualthumb', 'frameless', 'framed', 'thumbnail']),
+	horizAlign = new Set(['left', 'right', 'center', 'none']),
+	vertAlign = new Set(['baseline', 'sub', 'super', 'top', 'text-top', 'middle', 'bottom', 'text-bottom']);
+
+/**
+ * 图片
+ * @classdesc `{childNodes: [AtomToken, ...ImageParameterToken]}`
+ */
+class FileToken extends LinkToken {
+	type = 'file';
+
+	/**
+	 * @param {string} link 文件名
+	 * @param {string|undefined} text 图片参数
+	 * @param {accum} accum
+	 * @param {string} delimiter `|`
+	 * @complexity `n`
+	 */
+	constructor(link, text, config = Parser.getConfig(), accum = [], delimiter = '|') {
+		super(link, undefined, config, accum, delimiter);
+		this.append(...explode('-{', '}-', '|', text).map(part => new ImageParameterToken(part, config, accum)));
+	}
+
+	/**
+	 * @override
+	 * @param {number} start 起始位置
+	 */
+	lint(start = this.getAbsoluteIndex()) {
+		const errors = super.lint(start),
+			args = this.getAllArgs().filter(({childNodes}) => {
+				const visibleNodes = childNodes.filter(node => node.text().trim());
+				return visibleNodes.length !== 1 || visibleNodes[0].type !== 'arg';
+			}),
+			keys = [...new Set(args.map(({name}) => name))].filter(key => key !== 'invalid'),
+			frameKeys = keys.filter(key => frame.has(key)),
+			horizAlignKeys = keys.filter(key => horizAlign.has(key)),
+			vertAlignKeys = keys.filter(key => vertAlign.has(key));
+		if (args.length === keys.length
+			&& frameKeys.length < 2 && horizAlignKeys.length < 2 && vertAlignKeys.length < 2
+		) {
+			return errors;
+		}
+		const rect = {start, ...this.getRootNode().posFromIndex(start)};
+		for (const key of keys) {
+			let relevantArgs = args.filter(({name}) => name === key);
+			if (key === 'caption') {
+				relevantArgs = [
+					...relevantArgs.slice(0, -1).filter(arg => arg.text()),
+					relevantArgs.at(-1),
+				];
+			}
+			if (relevantArgs.length > 1) {
+				errors.push(...relevantArgs.map(
+					arg => generateForChild(arg, rect, Parser.msg('duplicated image $1 parameter', key)),
+				));
+			}
+		}
+		if (frameKeys.size > 1) {
+			errors.push(
+				...args.filter(({name}) => frame.has(name)).map(
+					arg => generateForChild(arg, rect, 'conflicting image $1 parameter', 'frame'),
+				),
+			);
+		}
+		if (horizAlignKeys.size > 1) {
+			errors.push(
+				...args.filter(({name}) => horizAlign.has(name))
+					.map(arg => generateForChild(arg, rect, 'conflicting image $1 parameter', 'horizontal-alignment')),
+			);
+		}
+		if (vertAlignKeys.size > 1) {
+			errors.push(
+				...args.filter(({name}) => vertAlign.has(name))
+					.map(arg => generateForChild(arg, rect, 'conflicting image $1 parameter', 'vertical-alignment')),
+			);
+		}
+		return errors;
+	}
+
+	/**
+	 * 获取所有图片参数节点
+	 * @returns {ImageParameterToken[]}
+	 */
+	getAllArgs() {
+		return this.childNodes.slice(1);
+	}
+
+	/**
+	 * 获取指定图片参数
+	 * @param {string} key 参数名
+	 * @complexity `n`
+	 */
+	getArgs(key) {
+		return this.getAllArgs().filter(({name}) => key === name);
+	}
+
+	/**
+	 * 获取特定类型的图片属性参数节点
+	 * @param {Set<string>} keys 接受的参数名
+	 * @param {type} type 类型名
+	 * @complexity `n`
+	 */
+	#getTypedArgs(keys, type) {
+		const args = this.getAllArgs().filter(({name}) => keys.has(name));
+		return args;
+	}
+
+	/** 获取图片框架属性参数节点 */
+	getFrameArgs() {
+		return this.#getTypedArgs(frame, '框架');
+	}
+
+	/** 获取图片水平对齐参数节点 */
+	getHorizAlignArgs() {
+		return this.#getTypedArgs(horizAlign, '水平对齐');
+	}
+
+	/** 获取图片垂直对齐参数节点 */
+	getVertAlignArgs() {
+		return this.#getTypedArgs(vertAlign, '垂直对齐');
+	}
+}
+
+module.exports = FileToken;

package/src/link/galleryImage.js

@@ -0,0 +1,62 @@
+'use strict';
+
+const {generateForSelf} = require('../../util/lint'),
+	Parser = require('../..'),
+	Token = require('..'),
+	FileToken = require('./file');
+
+/**
+ * 图片
+ * @classdesc `{childNodes: [AtomToken, ...ImageParameterToken]}`
+ */
+class GalleryImageToken extends FileToken {
+	type = 'gallery-image';
+	#invalid = false;
+
+	/**
+	 * @param {string} link 图片文件名
+	 * @param {string|undefined} text 图片参数
+	 * @param {accum} accum
+	 */
+	constructor(link, text, config = Parser.getConfig(), accum = []) {
+		let token;
+		if (text !== undefined) {
+			token = new Token(text, config, true, accum);
+			token.type = 'temp';
+			for (let n = 1; n < Parser.MAX_STAGE; n++) {
+				token.getAttribute('parseOnce')();
+			}
+			accum.splice(accum.indexOf(token), 1);
+		}
+		super(link, token?.toString(), config, accum);
+		this.setAttribute('bracket', false);
+	}
+
+	/**
+	 * @override
+	 */
+	afterBuild() {
+		const initImagemap = this.type === 'imagemap-image',
+			titleObj = this.normalizeTitle(String(this.firstChild), initImagemap ? 0 : 6, true, !initImagemap);
+		this.#invalid = titleObj.ns !== 6; // 只用于gallery-image的首次解析
+	}
+
+	/** @override */
+	getPadding() {
+		return 0;
+	}
+
+	/**
+	 * @override
+	 * @param {number} start 起始位置
+	 */
+	lint(start = this.getAbsoluteIndex()) {
+		const errors = super.lint(start);
+		if (this.#invalid) {
+			errors.push(generateForSelf(this, {start}, 'invalid gallery image'));
+		}
+		return errors;
+	}
+}
+
+module.exports = GalleryImageToken;