scrapex 0.5.3 → 1.0.0-beta.1

package/dist/parsers-CwkYnyWY.mjs

@@ -0,0 +1,482 @@
+import * as cheerio from "cheerio";
+import { fromMarkdown } from "mdast-util-from-markdown";
+import { toString } from "mdast-util-to-string";
+import { visit } from "unist-util-visit";
+
+//#region src/parsers/github.ts
+/**
+* GitHub-specific utilities for parsing repositories.
+*/
+/**
+* Check if a URL is a GitHub repository
+*/
+function isGitHubRepo(url) {
+	return /^https?:\/\/(www\.)?github\.com\/[^/]+\/[^/]+\/?$/.test(url);
+}
+/**
+* Extract GitHub repo info from URL
+*/
+function parseGitHubUrl(url) {
+	const match = url.match(/github\.com\/([^/]+)\/([^/]+)/);
+	if (!match || !match[1] || !match[2]) return null;
+	return {
+		owner: match[1],
+		repo: match[2].replace(/\.git$/, "")
+	};
+}
+/**
+* Convert a GitHub repo URL to raw content URL
+*/
+function toRawUrl(url, branch = "main", file = "README.md") {
+	const info = parseGitHubUrl(url);
+	if (!info) return url;
+	return `https://raw.githubusercontent.com/${info.owner}/${info.repo}/${branch}/${file}`;
+}
+/**
+* Fetch GitHub API metadata for a repository
+* Note: This is a placeholder - actual implementation would need GitHub API access
+*/
+async function fetchRepoMeta(owner, repo, _token) {
+	return {
+		repoOwner: owner,
+		repoName: repo
+	};
+}
+/**
+* Group links by their category/section
+*/
+function groupByCategory(links) {
+	const groups = /* @__PURE__ */ new Map();
+	for (const link of links) {
+		const category = link.context || "Uncategorized";
+		const existing = groups.get(category) || [];
+		existing.push(link);
+		groups.set(category, existing);
+	}
+	return groups;
+}
+
+//#endregion
+//#region src/parsers/markdown.ts
+/**
+* Generic Markdown parser.
+* Extracts structure, links, and code blocks from markdown content.
+*
+* @example
+* ```ts
+* const parser = new MarkdownParser();
+* const result = parser.parse(markdownContent);
+* console.log(result.data.sections);
+* console.log(result.data.links);
+* ```
+*/
+var MarkdownParser = class {
+	name = "markdown";
+	canParse(content) {
+		return content.includes("# ") || content.includes("## ") || content.includes("- [") || content.includes("* [") || content.includes("```");
+	}
+	parse(content) {
+		const tree = fromMarkdown(content);
+		const sections = [];
+		const allLinks = [];
+		const codeBlocks = [];
+		let frontmatter;
+		if (content.startsWith("---")) {
+			const endIndex = content.indexOf("---", 3);
+			if (endIndex !== -1) {
+				const frontmatterContent = content.slice(3, endIndex).trim();
+				frontmatter = this.parseFrontmatter(frontmatterContent);
+			}
+		}
+		let currentSection = null;
+		visit(tree, (node) => {
+			if (node.type === "heading") {
+				const heading = node;
+				const title = toString(heading);
+				if (currentSection) sections.push(currentSection);
+				currentSection = {
+					level: heading.depth,
+					title,
+					content: "",
+					links: []
+				};
+			}
+			if (node.type === "link") {
+				const link = node;
+				const text = toString(link);
+				const linkData = {
+					url: link.url,
+					text,
+					title: link.title ?? void 0,
+					context: currentSection?.title
+				};
+				allLinks.push(linkData);
+				if (currentSection) currentSection.links.push(linkData);
+			}
+			if (node.type === "code") {
+				const code = node;
+				codeBlocks.push({
+					language: code.lang ?? void 0,
+					code: code.value,
+					meta: code.meta ?? void 0
+				});
+			}
+			if (currentSection && node.type === "paragraph") {
+				const text = toString(node);
+				currentSection.content += (currentSection.content ? "\n\n" : "") + text;
+			}
+		});
+		if (currentSection) sections.push(currentSection);
+		return { data: {
+			title: frontmatter?.title ?? sections.find((s) => s.level === 1)?.title,
+			description: frontmatter?.description ?? this.extractDescription(tree),
+			sections,
+			links: allLinks,
+			codeBlocks,
+			frontmatter
+		} };
+	}
+	parseFrontmatter(content) {
+		const result = {};
+		const lines = content.split("\n");
+		for (const line of lines) {
+			const colonIndex = line.indexOf(":");
+			if (colonIndex > 0) {
+				const key = line.slice(0, colonIndex).trim();
+				let value = line.slice(colonIndex + 1).trim();
+				if (value === "true") value = true;
+				else if (value === "false") value = false;
+				else if (/^-?\d+(\.\d+)?$/.test(value)) value = Number(value);
+				else if (value.startsWith("\"") && value.endsWith("\"")) value = value.slice(1, -1);
+				else if (value.startsWith("'") && value.endsWith("'")) value = value.slice(1, -1);
+				result[key] = value;
+			}
+		}
+		return result;
+	}
+	extractDescription(tree) {
+		for (const node of tree.children) {
+			if (node.type === "heading") break;
+			if (node.type === "paragraph") return toString(node);
+		}
+	}
+};
+/**
+* Extract links from a list-based markdown structure (like awesome lists)
+*/
+function extractListLinks(markdown) {
+	const tree = fromMarkdown(markdown);
+	const links = [];
+	let currentHeading = "";
+	visit(tree, (node) => {
+		if (node.type === "heading") currentHeading = toString(node);
+		if (node.type === "listItem") visit(node, "link", (linkNode) => {
+			links.push({
+				url: linkNode.url,
+				text: toString(linkNode),
+				title: linkNode.title ?? void 0,
+				context: currentHeading || void 0
+			});
+		});
+	});
+	return links;
+}
+/**
+* Parse markdown into sections by heading level
+*/
+function parseByHeadings(markdown, minLevel = 2) {
+	return new MarkdownParser().parse(markdown).data.sections.filter((s) => s.level >= minLevel);
+}
+
+//#endregion
+//#region src/parsers/rss.ts
+/**
+* RSS/Atom feed parser.
+* Supports RSS 2.0, RSS 1.0 (RDF), and Atom 1.0 formats.
+*
+* @example
+* ```ts
+* import { RSSParser } from 'scrapex/parsers';
+*
+* const parser = new RSSParser();
+* const result = parser.parse(feedXml, 'https://example.com/feed.xml');
+*
+* console.log(result.data.title);
+* console.log(result.data.items);
+* ```
+*/
+var RSSParser = class {
+	name = "rss";
+	customFields;
+	constructor(options) {
+		this.customFields = options?.customFields || {};
+	}
+	canParse(content) {
+		const lower = content.toLowerCase();
+		return lower.includes("<rss") || lower.includes("<feed") || lower.includes("<rdf:rdf");
+	}
+	parse(content, url) {
+		const $ = cheerio.load(content, { xml: true });
+		if ($("feed").length > 0) return this.parseAtom($, url);
+		else if ($("rdf\\:RDF, RDF").length > 0) return this.parseRSS1($, url);
+		else return this.parseRSS2($, url);
+	}
+	parseRSS2($, baseUrl) {
+		const channel = $("channel");
+		const feedLink = channel.find("> link").text();
+		const resolveBase = baseUrl || feedLink;
+		const items = $("item").map((_, el) => {
+			const $item = $(el);
+			const itemLink = $item.find("link").text();
+			const guid = $item.find("guid").text();
+			const pubDate = $item.find("pubDate").text();
+			const resolvedLink = this.resolveLink(itemLink, guid, resolveBase);
+			return {
+				id: guid || itemLink,
+				title: $item.find("title").text(),
+				link: resolvedLink,
+				description: this.parseText($item.find("description")),
+				content: this.parseContentEncoded($item.find("content\\:encoded")),
+				author: $item.find("author").text() || $item.find("dc\\:creator").text() || void 0,
+				publishedAt: this.parseDate(pubDate),
+				rawPublishedAt: pubDate || void 0,
+				categories: this.parseCategories($item.find("category").map((_$1, c) => $(c).text()).get()),
+				enclosure: this.parseEnclosure($item.find("enclosure"), resolveBase),
+				customFields: this.extractCustomFields($item)
+			};
+		}).get();
+		return {
+			data: {
+				format: "rss2",
+				title: channel.find("> title").text(),
+				description: channel.find("> description").text() || void 0,
+				link: this.resolveUrl(feedLink, resolveBase),
+				language: channel.find("> language").text() || void 0,
+				lastBuildDate: this.parseDate(channel.find("> lastBuildDate").text()),
+				copyright: channel.find("> copyright").text() || void 0,
+				items,
+				customFields: this.extractCustomFields(channel)
+			},
+			meta: {
+				generator: channel.find("> generator").text() || void 0,
+				ttl: this.parseNumber(channel.find("> ttl").text()),
+				image: this.parseImage(channel.find("> image"), resolveBase),
+				categories: this.parseCategories(channel.find("> category").map((_, c) => $(c).text()).get())
+			}
+		};
+	}
+	parseAtom($, baseUrl) {
+		const feed = $("feed");
+		const feedLink = feed.find("> link[rel=\"alternate\"], > link:not([rel])").attr("href") || "";
+		const nextLink = feed.find("> link[rel=\"next\"]").attr("href");
+		const resolveBase = baseUrl || feedLink;
+		const items = $("entry").map((_, el) => {
+			const $entry = $(el);
+			const entryLink = $entry.find("link[rel=\"alternate\"], link:not([rel])").attr("href") || "";
+			const entryId = $entry.find("id").text();
+			const published = $entry.find("published").text();
+			const updated = $entry.find("updated").text();
+			const resolvedLink = this.resolveLink(entryLink, entryId, resolveBase);
+			return {
+				id: entryId,
+				title: $entry.find("title").text(),
+				link: resolvedLink,
+				description: this.parseText($entry.find("summary")),
+				content: this.parseText($entry.find("content")),
+				author: $entry.find("author name").text() || void 0,
+				publishedAt: this.parseDate(published),
+				rawPublishedAt: published || updated || void 0,
+				updatedAt: this.parseDate(updated),
+				categories: this.parseCategories($entry.find("category").map((_$1, c) => $(c).attr("term")).get()),
+				customFields: this.extractCustomFields($entry)
+			};
+		}).get();
+		return {
+			data: {
+				format: "atom",
+				title: feed.find("> title").text(),
+				description: feed.find("> subtitle").text() || void 0,
+				link: this.resolveUrl(feedLink, resolveBase),
+				next: nextLink ? this.resolveUrl(nextLink, resolveBase) : void 0,
+				language: feed.attr("xml:lang") || void 0,
+				lastBuildDate: this.parseDate(feed.find("> updated").text()),
+				copyright: feed.find("> rights").text() || void 0,
+				items,
+				customFields: this.extractCustomFields(feed)
+			},
+			meta: {
+				generator: feed.find("> generator").text() || void 0,
+				image: this.parseAtomImage(feed, resolveBase),
+				categories: this.parseCategories(feed.find("> category").map((_, c) => $(c).attr("term")).get())
+			}
+		};
+	}
+	parseRSS1($, baseUrl) {
+		const channel = $("channel");
+		const feedLink = channel.find("link").text();
+		const resolveBase = baseUrl || feedLink;
+		const items = $("item").map((_, el) => {
+			const $item = $(el);
+			const itemLink = $item.find("link").text();
+			const rdfAbout = $item.attr("rdf:about") || "";
+			const dcDate = $item.find("dc\\:date").text();
+			const resolvedLink = this.resolveLink(itemLink, rdfAbout, resolveBase);
+			const dcSubjects = $item.find("dc\\:subject").map((_$1, s) => $(s).text()).get();
+			return {
+				id: rdfAbout || itemLink,
+				title: $item.find("title").text(),
+				link: resolvedLink,
+				description: this.parseText($item.find("description")),
+				content: this.parseContentEncoded($item.find("content\\:encoded")),
+				author: $item.find("dc\\:creator").text() || void 0,
+				publishedAt: this.parseDate(dcDate),
+				rawPublishedAt: dcDate || void 0,
+				categories: this.parseCategories(dcSubjects),
+				customFields: this.extractCustomFields($item)
+			};
+		}).get();
+		const rdfImage = $("image");
+		const imageUrl = rdfImage.find("url").text() || rdfImage.attr("rdf:resource");
+		return {
+			data: {
+				format: "rss1",
+				title: channel.find("title").text(),
+				description: channel.find("description").text() || void 0,
+				link: this.resolveUrl(feedLink, resolveBase),
+				language: channel.find("dc\\:language").text() || void 0,
+				lastBuildDate: this.parseDate(channel.find("dc\\:date").text()),
+				copyright: channel.find("dc\\:rights").text() || void 0,
+				items,
+				customFields: this.extractCustomFields(channel)
+			},
+			meta: {
+				generator: channel.find("admin\\:generatorAgent").attr("rdf:resource") || void 0,
+				image: imageUrl ? {
+					url: this.resolveUrl(imageUrl, resolveBase),
+					title: rdfImage.find("title").text() || void 0,
+					link: this.resolveUrl(rdfImage.find("link").text(), resolveBase) || void 0
+				} : void 0,
+				categories: this.parseCategories(channel.find("dc\\:subject").map((_, s) => $(s).text()).get())
+			}
+		};
+	}
+	/**
+	* Extract custom fields from an element using configured selectors.
+	*/
+	extractCustomFields($el) {
+		if (Object.keys(this.customFields).length === 0) return void 0;
+		const fields = {};
+		for (const [key, selector] of Object.entries(this.customFields)) {
+			const value = $el.find(selector).text().trim();
+			if (value) fields[key] = value;
+		}
+		return Object.keys(fields).length > 0 ? fields : void 0;
+	}
+	/**
+	* Parse date string to ISO 8601 format.
+	* Returns undefined if parsing fails (never returns raw strings).
+	*/
+	parseDate(dateStr) {
+		if (!dateStr?.trim()) return void 0;
+		try {
+			const date = new Date(dateStr);
+			if (Number.isNaN(date.getTime())) return;
+			return date.toISOString();
+		} catch {
+			return;
+		}
+	}
+	/**
+	* Parse element text content, returning undefined if empty.
+	*/
+	parseText($el) {
+		return $el.text().trim() || void 0;
+	}
+	/**
+	* Parse content:encoded, stripping HTML tags from CDATA content.
+	*/
+	parseContentEncoded($el) {
+		const raw = $el.text().trim();
+		if (!raw) return void 0;
+		return raw.replace(/<[^>]+>/g, "").trim() || void 0;
+	}
+	/**
+	* Parse categories/tags, filtering out empty strings.
+	*/
+	parseCategories(categories) {
+		return categories.map((c) => c?.trim()).filter((c) => !!c && c.length > 0);
+	}
+	/**
+	* Resolve a URL against a base URL.
+	* Only allows https scheme; all other schemes are rejected.
+	*/
+	resolveUrl(url, base) {
+		if (!url?.trim()) return "";
+		try {
+			const resolved = base ? new URL(url, base) : new URL(url);
+			return resolved.protocol === "https:" ? resolved.href : "";
+		} catch {
+			return "";
+		}
+	}
+	/**
+	* Resolve link with fallback to id/guid if primary link is empty and id is a URL.
+	* Only allows https scheme; all other schemes are rejected.
+	*/
+	resolveLink(primaryLink, fallbackId, base) {
+		if (primaryLink?.trim()) return this.resolveUrl(primaryLink, base);
+		if (fallbackId?.trim()) try {
+			const resolvedId = new URL(fallbackId);
+			return resolvedId.protocol === "https:" ? resolvedId.href : "";
+		} catch {
+			return this.resolveUrl(fallbackId, base);
+		}
+		return "";
+	}
+	/**
+	* Parse enclosure element with URL resolution.
+	*/
+	parseEnclosure($enc, baseUrl) {
+		const url = $enc.attr("url");
+		if (!url) return void 0;
+		return {
+			url: this.resolveUrl(url, baseUrl),
+			type: $enc.attr("type") || void 0,
+			length: this.parseNumber($enc.attr("length"))
+		};
+	}
+	/**
+	* Parse RSS image element with URL resolution.
+	*/
+	parseImage($img, baseUrl) {
+		const url = $img.find("url").text();
+		if (!url) return void 0;
+		return {
+			url: this.resolveUrl(url, baseUrl),
+			title: $img.find("title").text() || void 0,
+			link: this.resolveUrl($img.find("link").text(), baseUrl) || void 0
+		};
+	}
+	/**
+	* Parse Atom logo/icon element with URL resolution.
+	*/
+	parseAtomImage($feed, baseUrl) {
+		const logo = $feed.find("> logo").text();
+		const icon = $feed.find("> icon").text();
+		const url = logo || icon;
+		if (!url) return void 0;
+		return { url: this.resolveUrl(url, baseUrl) };
+	}
+	/**
+	* Parse a string to number, returning undefined if invalid.
+	*/
+	parseNumber(value) {
+		if (!value) return void 0;
+		const num = parseInt(value, 10);
+		return Number.isNaN(num) ? void 0 : num;
+	}
+};
+
+//#endregion
+export { fetchRepoMeta as a, parseGitHubUrl as c, parseByHeadings as i, toRawUrl as l, MarkdownParser as n, groupByCategory as o, extractListLinks as r, isGitHubRepo as s, RSSParser as t };
+//# sourceMappingURL=parsers-CwkYnyWY.mjs.map

package/dist/parsers-CwkYnyWY.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"parsers-CwkYnyWY.mjs","names":["sections: MarkdownSection[]","allLinks: MarkdownLink[]","codeBlocks: CodeBlock[]","frontmatter: Record<string, unknown> | undefined","currentSection: MarkdownSection | null","mdastToString","linkData: MarkdownLink","result: Record<string, unknown>","value: string | boolean | number","links: MarkdownLink[]","items: FeedItem[]","fields: Record<string, string>"],"sources":["../src/parsers/github.ts","../src/parsers/markdown.ts","../src/parsers/rss.ts"],"sourcesContent":["import type { GitHubMeta, MarkdownLink } from './types.js';\n\n/**\n * GitHub-specific utilities for parsing repositories.\n */\n\n/**\n * Check if a URL is a GitHub repository\n */\nexport function isGitHubRepo(url: string): boolean {\n  return /^https?:\\/\\/(www\\.)?github\\.com\\/[^/]+\\/[^/]+\\/?$/.test(url);\n}\n\n/**\n * Extract GitHub repo info from URL\n */\nexport function parseGitHubUrl(url: string): { owner: string; repo: string } | null {\n  const match = url.match(/github\\.com\\/([^/]+)\\/([^/]+)/);\n  if (!match || !match[1] || !match[2]) return null;\n  return {\n    owner: match[1],\n    repo: match[2].replace(/\\.git$/, ''),\n  };\n}\n\n/**\n * Convert a GitHub repo URL to raw content URL\n */\nexport function toRawUrl(url: string, branch = 'main', file = 'README.md'): string {\n  const info = parseGitHubUrl(url);\n  if (!info) return url;\n  return `https://raw.githubusercontent.com/${info.owner}/${info.repo}/${branch}/${file}`;\n}\n\n/**\n * Fetch GitHub API metadata for a repository\n * Note: This is a placeholder - actual implementation would need GitHub API access\n */\nexport async function fetchRepoMeta(\n  owner: string,\n  repo: string,\n  _token?: string\n): Promise<GitHubMeta> {\n  // This would make actual API calls in a full implementation\n  // For now, return basic info\n  return {\n    repoOwner: owner,\n    repoName: repo,\n  };\n}\n\n/**\n * Group links by their category/section\n */\nexport function groupByCategory(links: MarkdownLink[]): Map<string, MarkdownLink[]> {\n  const groups = new Map<string, MarkdownLink[]>();\n\n  for (const link of links) {\n    const category = link.context || 'Uncategorized';\n    const existing = groups.get(category) || [];\n    existing.push(link);\n    groups.set(category, existing);\n  }\n\n  return groups;\n}\n","import type { Code, Heading, Link, ListItem, Root } from 'mdast';\nimport { fromMarkdown } from 'mdast-util-from-markdown';\nimport { toString as mdastToString } from 'mdast-util-to-string';\nimport { visit } from 'unist-util-visit';\nimport type {\n  CodeBlock,\n  MarkdownLink,\n  MarkdownSection,\n  ParsedMarkdown,\n  ParserResult,\n  SourceParser,\n} from './types.js';\n\n/**\n * Generic Markdown parser.\n * Extracts structure, links, and code blocks from markdown content.\n *\n * @example\n * ```ts\n * const parser = new MarkdownParser();\n * const result = parser.parse(markdownContent);\n * console.log(result.data.sections);\n * console.log(result.data.links);\n * ```\n */\nexport class MarkdownParser implements SourceParser<ParsedMarkdown> {\n  readonly name = 'markdown';\n\n  canParse(content: string): boolean {\n    // Check for common markdown patterns\n    return (\n      content.includes('# ') ||\n      content.includes('## ') ||\n      content.includes('- [') ||\n      content.includes('* [') ||\n      content.includes('```')\n    );\n  }\n\n  parse(content: string): ParserResult<ParsedMarkdown> {\n    const tree = fromMarkdown(content);\n    const sections: MarkdownSection[] = [];\n    const allLinks: MarkdownLink[] = [];\n    const codeBlocks: CodeBlock[] = [];\n    let frontmatter: Record<string, unknown> | undefined;\n\n    // Extract frontmatter if present\n    if (content.startsWith('---')) {\n      const endIndex = content.indexOf('---', 3);\n      if (endIndex !== -1) {\n        const frontmatterContent = content.slice(3, endIndex).trim();\n        frontmatter = this.parseFrontmatter(frontmatterContent);\n      }\n    }\n\n    // Track current section\n    let currentSection: MarkdownSection | null = null;\n\n    // Process the AST\n    visit(tree, (node) => {\n      // Handle headings\n      if (node.type === 'heading') {\n        const heading = node as Heading;\n        const title = mdastToString(heading);\n\n        // Finalize previous section\n        if (currentSection) {\n          sections.push(currentSection);\n        }\n\n        currentSection = {\n          level: heading.depth,\n          title,\n          content: '',\n          links: [],\n        };\n      }\n\n      // Handle links\n      if (node.type === 'link') {\n        const link = node as Link;\n        const text = mdastToString(link);\n        const linkData: MarkdownLink = {\n          url: link.url,\n          text,\n          title: link.title ?? undefined,\n          context: currentSection?.title,\n        };\n\n        allLinks.push(linkData);\n        if (currentSection) {\n          currentSection.links.push(linkData);\n        }\n      }\n\n      // Handle code blocks\n      if (node.type === 'code') {\n        const code = node as Code;\n        codeBlocks.push({\n          language: code.lang ?? undefined,\n          code: code.value,\n          meta: code.meta ?? undefined,\n        });\n      }\n\n      // Accumulate content for current section\n      if (currentSection && node.type === 'paragraph') {\n        const text = mdastToString(node);\n        currentSection.content += (currentSection.content ? '\\n\\n' : '') + text;\n      }\n    });\n\n    // Finalize last section\n    if (currentSection) {\n      sections.push(currentSection);\n    }\n\n    // Extract title from first h1 or frontmatter\n    const title = (frontmatter?.title as string) ?? sections.find((s) => s.level === 1)?.title;\n\n    // Extract description from frontmatter or first paragraph before any heading\n    const description = (frontmatter?.description as string) ?? this.extractDescription(tree);\n\n    return {\n      data: {\n        title,\n        description,\n        sections,\n        links: allLinks,\n        codeBlocks,\n        frontmatter,\n      },\n    };\n  }\n\n  private parseFrontmatter(content: string): Record<string, unknown> {\n    const result: Record<string, unknown> = {};\n    const lines = content.split('\\n');\n\n    for (const line of lines) {\n      const colonIndex = line.indexOf(':');\n      if (colonIndex > 0) {\n        const key = line.slice(0, colonIndex).trim();\n        let value: string | boolean | number = line.slice(colonIndex + 1).trim();\n\n        // Parse simple types\n        if (value === 'true') value = true;\n        else if (value === 'false') value = false;\n        else if (/^-?\\d+(\\.\\d+)?$/.test(value)) value = Number(value);\n        else if (value.startsWith('\"') && value.endsWith('\"')) value = value.slice(1, -1);\n        else if (value.startsWith(\"'\") && value.endsWith(\"'\")) value = value.slice(1, -1);\n\n        result[key] = value;\n      }\n    }\n\n    return result;\n  }\n\n  private extractDescription(tree: Root): string | undefined {\n    // Find first paragraph before any heading\n    for (const node of tree.children) {\n      if (node.type === 'heading') break;\n      if (node.type === 'paragraph') {\n        return mdastToString(node);\n      }\n    }\n    return undefined;\n  }\n}\n\n/**\n * Extract links from a list-based markdown structure (like awesome lists)\n */\nexport function extractListLinks(markdown: string): MarkdownLink[] {\n  const tree = fromMarkdown(markdown);\n  const links: MarkdownLink[] = [];\n  let currentHeading = '';\n\n  visit(tree, (node) => {\n    if (node.type === 'heading') {\n      currentHeading = mdastToString(node as Heading);\n    }\n\n    if (node.type === 'listItem') {\n      const listItem = node as ListItem;\n\n      // Find links in this list item\n      visit(listItem, 'link', (linkNode: Link) => {\n        links.push({\n          url: linkNode.url,\n          text: mdastToString(linkNode),\n          title: linkNode.title ?? undefined,\n          context: currentHeading || undefined,\n        });\n      });\n    }\n  });\n\n  return links;\n}\n\n/**\n * Parse markdown into sections by heading level\n */\nexport function parseByHeadings(markdown: string, minLevel = 2): MarkdownSection[] {\n  const parser = new MarkdownParser();\n  const result = parser.parse(markdown);\n  return result.data.sections.filter((s) => s.level >= minLevel);\n}\n","import * as cheerio from 'cheerio';\nimport type { Element } from 'domhandler';\nimport type {\n  FeedEnclosure,\n  FeedItem,\n  FeedMeta,\n  ParsedFeed,\n  ParserResult,\n  SourceParser,\n} from './types.js';\n\nexport interface RSSParserOptions {\n  /**\n   * Map of custom field names to CSS selectors or XML tag names.\n   * Useful for extracting podcast/media namespace fields.\n   * @example { duration: \"itunes\\:duration\", rating: \"media\\:rating\" }\n   */\n  customFields?: Record<string, string>;\n}\n\n/**\n * RSS/Atom feed parser.\n * Supports RSS 2.0, RSS 1.0 (RDF), and Atom 1.0 formats.\n *\n * @example\n * ```ts\n * import { RSSParser } from 'scrapex/parsers';\n *\n * const parser = new RSSParser();\n * const result = parser.parse(feedXml, 'https://example.com/feed.xml');\n *\n * console.log(result.data.title);\n * console.log(result.data.items);\n * ```\n */\nexport class RSSParser implements SourceParser<ParsedFeed, FeedMeta> {\n  readonly name = 'rss';\n  private customFields: Record<string, string>;\n\n  constructor(options?: RSSParserOptions) {\n    this.customFields = options?.customFields || {};\n  }\n\n  canParse(content: string): boolean {\n    const lower = content.toLowerCase();\n    return lower.includes('<rss') || lower.includes('<feed') || lower.includes('<rdf:rdf');\n  }\n\n  parse(content: string, url?: string): ParserResult<ParsedFeed, FeedMeta> {\n    // Cheerio's xml: true mode disables HTML entities and structure fixes,\n    // effectively preventing many XSS/injection vectors during parsing.\n    const $ = cheerio.load(content, { xml: true });\n\n    // Detect format and parse\n    if ($('feed').length > 0) {\n      return this.parseAtom($, url);\n    } else if ($('rdf\\\\:RDF, RDF').length > 0) {\n      return this.parseRSS1($, url);\n    } else {\n      return this.parseRSS2($, url);\n    }\n  }\n\n  private parseRSS2($: cheerio.CheerioAPI, baseUrl?: string): ParserResult<ParsedFeed, FeedMeta> {\n    const channel = $('channel');\n    const feedLink = channel.find('> link').text();\n    const resolveBase = baseUrl || feedLink;\n\n    const items: FeedItem[] = $('item')\n      .map((_, el) => {\n        const $item = $(el);\n        const itemLink = $item.find('link').text();\n        const guid = $item.find('guid').text();\n        const pubDate = $item.find('pubDate').text();\n\n        // Resolve link with fallback to guid if it's a URL\n        const resolvedLink = this.resolveLink(itemLink, guid, resolveBase);\n\n        return {\n          id: guid || itemLink,\n          title: $item.find('title').text(),\n          link: resolvedLink,\n          description: this.parseText($item.find('description')),\n          content: this.parseContentEncoded($item.find('content\\\\:encoded')),\n          author: $item.find('author').text() || $item.find('dc\\\\:creator').text() || undefined,\n          publishedAt: this.parseDate(pubDate),\n          rawPublishedAt: pubDate || undefined,\n          categories: this.parseCategories(\n            $item\n              .find('category')\n              .map((_, c) => $(c).text())\n              .get()\n          ),\n          enclosure: this.parseEnclosure($item.find('enclosure'), resolveBase),\n          customFields: this.extractCustomFields($item),\n        };\n      })\n      .get();\n\n    return {\n      data: {\n        format: 'rss2',\n        title: channel.find('> title').text(),\n        description: channel.find('> description').text() || undefined,\n        link: this.resolveUrl(feedLink, resolveBase),\n        language: channel.find('> language').text() || undefined,\n        lastBuildDate: this.parseDate(channel.find('> lastBuildDate').text()),\n        copyright: channel.find('> copyright').text() || undefined,\n        items,\n        customFields: this.extractCustomFields(channel),\n      },\n      meta: {\n        generator: channel.find('> generator').text() || undefined,\n        ttl: this.parseNumber(channel.find('> ttl').text()),\n        image: this.parseImage(channel.find('> image'), resolveBase),\n        categories: this.parseCategories(\n          channel\n            .find('> category')\n            .map((_, c) => $(c).text())\n            .get()\n        ),\n      },\n    };\n  }\n\n  private parseAtom($: cheerio.CheerioAPI, baseUrl?: string): ParserResult<ParsedFeed, FeedMeta> {\n    const feed = $('feed');\n    const feedLink = feed.find('> link[rel=\"alternate\"], > link:not([rel])').attr('href') || '';\n    const nextLink = feed.find('> link[rel=\"next\"]').attr('href');\n    const resolveBase = baseUrl || feedLink;\n\n    const items: FeedItem[] = $('entry')\n      .map((_, el) => {\n        const $entry = $(el);\n        const entryLink = $entry.find('link[rel=\"alternate\"], link:not([rel])').attr('href') || '';\n        const entryId = $entry.find('id').text();\n        const published = $entry.find('published').text();\n        const updated = $entry.find('updated').text();\n\n        // Resolve link with fallback to id if it's a URL\n        const resolvedLink = this.resolveLink(entryLink, entryId, resolveBase);\n\n        return {\n          id: entryId,\n          title: $entry.find('title').text(),\n          link: resolvedLink,\n          description: this.parseText($entry.find('summary')),\n          content: this.parseText($entry.find('content')),\n          author: $entry.find('author name').text() || undefined,\n          publishedAt: this.parseDate(published),\n          rawPublishedAt: published || updated || undefined,\n          updatedAt: this.parseDate(updated),\n          categories: this.parseCategories(\n            $entry\n              .find('category')\n              .map((_, c) => $(c).attr('term'))\n              .get()\n          ),\n          customFields: this.extractCustomFields($entry),\n        };\n      })\n      .get();\n\n    return {\n      data: {\n        format: 'atom',\n        title: feed.find('> title').text(),\n        description: feed.find('> subtitle').text() || undefined,\n        link: this.resolveUrl(feedLink, resolveBase),\n        next: nextLink ? this.resolveUrl(nextLink, resolveBase) : undefined,\n        language: feed.attr('xml:lang') || undefined,\n        lastBuildDate: this.parseDate(feed.find('> updated').text()),\n        copyright: feed.find('> rights').text() || undefined,\n        items,\n        customFields: this.extractCustomFields(feed),\n      },\n      meta: {\n        generator: feed.find('> generator').text() || undefined,\n        image: this.parseAtomImage(feed, resolveBase),\n        categories: this.parseCategories(\n          feed\n            .find('> category')\n            .map((_, c) => $(c).attr('term'))\n            .get()\n        ),\n      },\n    };\n  }\n\n  private parseRSS1($: cheerio.CheerioAPI, baseUrl?: string): ParserResult<ParsedFeed, FeedMeta> {\n    const channel = $('channel');\n    const feedLink = channel.find('link').text();\n    const resolveBase = baseUrl || feedLink;\n\n    // RSS 1.0 items are siblings of channel, not children\n    const items: FeedItem[] = $('item')\n      .map((_, el) => {\n        const $item = $(el);\n        const itemLink = $item.find('link').text();\n        const rdfAbout = $item.attr('rdf:about') || '';\n        const dcDate = $item.find('dc\\\\:date').text();\n\n        // Resolve link with fallback to rdf:about\n        const resolvedLink = this.resolveLink(itemLink, rdfAbout, resolveBase);\n\n        // Extract dc:subject as categories (RSS 1.0 uses Dublin Core)\n        const dcSubjects = $item\n          .find('dc\\\\:subject')\n          .map((_, s) => $(s).text())\n          .get();\n\n        return {\n          id: rdfAbout || itemLink,\n          title: $item.find('title').text(),\n          link: resolvedLink,\n          description: this.parseText($item.find('description')),\n          content: this.parseContentEncoded($item.find('content\\\\:encoded')),\n          author: $item.find('dc\\\\:creator').text() || undefined,\n          publishedAt: this.parseDate(dcDate),\n          rawPublishedAt: dcDate || undefined,\n          categories: this.parseCategories(dcSubjects),\n          customFields: this.extractCustomFields($item),\n        };\n      })\n      .get();\n\n    // Parse RDF image element (sibling of channel)\n    const rdfImage = $('image');\n    const imageUrl = rdfImage.find('url').text() || rdfImage.attr('rdf:resource');\n\n    return {\n      data: {\n        format: 'rss1',\n        title: channel.find('title').text(),\n        description: channel.find('description').text() || undefined,\n        link: this.resolveUrl(feedLink, resolveBase),\n        language: channel.find('dc\\\\:language').text() || undefined,\n        lastBuildDate: this.parseDate(channel.find('dc\\\\:date').text()),\n        copyright: channel.find('dc\\\\:rights').text() || undefined,\n        items,\n        customFields: this.extractCustomFields(channel),\n      },\n      meta: {\n        generator: channel.find('admin\\\\:generatorAgent').attr('rdf:resource') || undefined,\n        image: imageUrl\n          ? {\n              url: this.resolveUrl(imageUrl, resolveBase),\n              title: rdfImage.find('title').text() || undefined,\n              link: this.resolveUrl(rdfImage.find('link').text(), resolveBase) || undefined,\n            }\n          : undefined,\n        categories: this.parseCategories(\n          channel\n            .find('dc\\\\:subject')\n            .map((_, s) => $(s).text())\n            .get()\n        ),\n      },\n    };\n  }\n\n  /**\n   * Extract custom fields from an element using configured selectors.\n   */\n  private extractCustomFields($el: cheerio.Cheerio<Element>): Record<string, string> | undefined {\n    if (Object.keys(this.customFields).length === 0) return undefined;\n\n    const fields: Record<string, string> = {};\n    for (const [key, selector] of Object.entries(this.customFields)) {\n      const value = $el.find(selector).text().trim();\n      if (value) {\n        fields[key] = value;\n      }\n    }\n\n    return Object.keys(fields).length > 0 ? fields : undefined;\n  }\n\n  /**\n   * Parse date string to ISO 8601 format.\n   * Returns undefined if parsing fails (never returns raw strings).\n   */\n  private parseDate(dateStr: string): string | undefined {\n    if (!dateStr?.trim()) return undefined;\n\n    try {\n      const date = new Date(dateStr);\n      // Check for invalid date\n      if (Number.isNaN(date.getTime())) {\n        return undefined;\n      }\n      return date.toISOString();\n    } catch {\n      return undefined;\n    }\n  }\n\n  /**\n   * Parse element text content, returning undefined if empty.\n   */\n  private parseText($el: cheerio.Cheerio<Element>): string | undefined {\n    const text = $el.text().trim();\n    return text || undefined;\n  }\n\n  /**\n   * Parse content:encoded, stripping HTML tags from CDATA content.\n   */\n  private parseContentEncoded($el: cheerio.Cheerio<Element>): string | undefined {\n    const raw = $el.text().trim();\n    if (!raw) return undefined;\n    return raw.replace(/<[^>]+>/g, '').trim() || undefined;\n  }\n\n  /**\n   * Parse categories/tags, filtering out empty strings.\n   */\n  private parseCategories(categories: (string | undefined)[]): string[] {\n    return categories.map((c) => c?.trim()).filter((c): c is string => !!c && c.length > 0);\n  }\n\n  /**\n   * Resolve a URL against a base URL.\n   * Only allows https scheme; all other schemes are rejected.\n   */\n  private resolveUrl(url: string, base?: string): string {\n    if (!url?.trim()) return '';\n\n    try {\n      const resolved = base ? new URL(url, base) : new URL(url);\n      return resolved.protocol === 'https:' ? resolved.href : '';\n    } catch {\n      return '';\n    }\n  }\n\n  /**\n   * Resolve link with fallback to id/guid if primary link is empty and id is a URL.\n   * Only allows https scheme; all other schemes are rejected.\n   */\n  private resolveLink(primaryLink: string, fallbackId: string, base?: string): string {\n    // Try primary link first\n    if (primaryLink?.trim()) {\n      return this.resolveUrl(primaryLink, base);\n    }\n\n    // Fallback to id if it looks like a URL\n    if (fallbackId?.trim()) {\n      try {\n        const resolvedId = new URL(fallbackId);\n        return resolvedId.protocol === 'https:' ? resolvedId.href : '';\n      } catch {\n        // Not a valid URL, try resolving against base\n        return this.resolveUrl(fallbackId, base);\n      }\n    }\n\n    return '';\n  }\n\n  /**\n   * Parse enclosure element with URL resolution.\n   */\n  private parseEnclosure(\n    $enc: cheerio.Cheerio<Element>,\n    baseUrl?: string\n  ): FeedEnclosure | undefined {\n    const url = $enc.attr('url');\n    if (!url) return undefined;\n\n    return {\n      url: this.resolveUrl(url, baseUrl),\n      type: $enc.attr('type') || undefined,\n      length: this.parseNumber($enc.attr('length')),\n    };\n  }\n\n  /**\n   * Parse RSS image element with URL resolution.\n   */\n  private parseImage(\n    $img: cheerio.Cheerio<Element>,\n    baseUrl?: string\n  ): FeedMeta['image'] | undefined {\n    const url = $img.find('url').text();\n    if (!url) return undefined;\n\n    return {\n      url: this.resolveUrl(url, baseUrl),\n      title: $img.find('title').text() || undefined,\n      link: this.resolveUrl($img.find('link').text(), baseUrl) || undefined,\n    };\n  }\n\n  /**\n   * Parse Atom logo/icon element with URL resolution.\n   */\n  private parseAtomImage(\n    $feed: cheerio.Cheerio<Element>,\n    baseUrl?: string\n  ): FeedMeta['image'] | undefined {\n    const logo = $feed.find('> logo').text();\n    const icon = $feed.find('> icon').text();\n    const url = logo || icon;\n\n    if (!url) return undefined;\n\n    return {\n      url: this.resolveUrl(url, baseUrl),\n    };\n  }\n\n  /**\n   * Parse a string to number, returning undefined if invalid.\n   */\n  private parseNumber(value: string | undefined): number | undefined {\n    if (!value) return undefined;\n    const num = parseInt(value, 10);\n    return Number.isNaN(num) ? undefined : num;\n  }\n}\n"],"mappings":";;;;;;;;;;;;AASA,SAAgB,aAAa,KAAsB;AACjD,QAAO,oDAAoD,KAAK,IAAI;;;;;AAMtE,SAAgB,eAAe,KAAqD;CAClF,MAAM,QAAQ,IAAI,MAAM,gCAAgC;AACxD,KAAI,CAAC,SAAS,CAAC,MAAM,MAAM,CAAC,MAAM,GAAI,QAAO;AAC7C,QAAO;EACL,OAAO,MAAM;EACb,MAAM,MAAM,GAAG,QAAQ,UAAU,GAAG;EACrC;;;;;AAMH,SAAgB,SAAS,KAAa,SAAS,QAAQ,OAAO,aAAqB;CACjF,MAAM,OAAO,eAAe,IAAI;AAChC,KAAI,CAAC,KAAM,QAAO;AAClB,QAAO,qCAAqC,KAAK,MAAM,GAAG,KAAK,KAAK,GAAG,OAAO,GAAG;;;;;;AAOnF,eAAsB,cACpB,OACA,MACA,QACqB;AAGrB,QAAO;EACL,WAAW;EACX,UAAU;EACX;;;;;AAMH,SAAgB,gBAAgB,OAAoD;CAClF,MAAM,yBAAS,IAAI,KAA6B;AAEhD,MAAK,MAAM,QAAQ,OAAO;EACxB,MAAM,WAAW,KAAK,WAAW;EACjC,MAAM,WAAW,OAAO,IAAI,SAAS,IAAI,EAAE;AAC3C,WAAS,KAAK,KAAK;AACnB,SAAO,IAAI,UAAU,SAAS;;AAGhC,QAAO;;;;;;;;;;;;;;;;;ACvCT,IAAa,iBAAb,MAAoE;CAClE,AAAS,OAAO;CAEhB,SAAS,SAA0B;AAEjC,SACE,QAAQ,SAAS,KAAK,IACtB,QAAQ,SAAS,MAAM,IACvB,QAAQ,SAAS,MAAM,IACvB,QAAQ,SAAS,MAAM,IACvB,QAAQ,SAAS,MAAM;;CAI3B,MAAM,SAA+C;EACnD,MAAM,OAAO,aAAa,QAAQ;EAClC,MAAMA,WAA8B,EAAE;EACtC,MAAMC,WAA2B,EAAE;EACnC,MAAMC,aAA0B,EAAE;EAClC,IAAIC;AAGJ,MAAI,QAAQ,WAAW,MAAM,EAAE;GAC7B,MAAM,WAAW,QAAQ,QAAQ,OAAO,EAAE;AAC1C,OAAI,aAAa,IAAI;IACnB,MAAM,qBAAqB,QAAQ,MAAM,GAAG,SAAS,CAAC,MAAM;AAC5D,kBAAc,KAAK,iBAAiB,mBAAmB;;;EAK3D,IAAIC,iBAAyC;AAG7C,QAAM,OAAO,SAAS;AAEpB,OAAI,KAAK,SAAS,WAAW;IAC3B,MAAM,UAAU;IAChB,MAAM,QAAQC,SAAc,QAAQ;AAGpC,QAAI,eACF,UAAS,KAAK,eAAe;AAG/B,qBAAiB;KACf,OAAO,QAAQ;KACf;KACA,SAAS;KACT,OAAO,EAAE;KACV;;AAIH,OAAI,KAAK,SAAS,QAAQ;IACxB,MAAM,OAAO;IACb,MAAM,OAAOA,SAAc,KAAK;IAChC,MAAMC,WAAyB;KAC7B,KAAK,KAAK;KACV;KACA,OAAO,KAAK,SAAS;KACrB,SAAS,gBAAgB;KAC1B;AAED,aAAS,KAAK,SAAS;AACvB,QAAI,eACF,gBAAe,MAAM,KAAK,SAAS;;AAKvC,OAAI,KAAK,SAAS,QAAQ;IACxB,MAAM,OAAO;AACb,eAAW,KAAK;KACd,UAAU,KAAK,QAAQ;KACvB,MAAM,KAAK;KACX,MAAM,KAAK,QAAQ;KACpB,CAAC;;AAIJ,OAAI,kBAAkB,KAAK,SAAS,aAAa;IAC/C,MAAM,OAAOD,SAAc,KAAK;AAChC,mBAAe,YAAY,eAAe,UAAU,SAAS,MAAM;;IAErE;AAGF,MAAI,eACF,UAAS,KAAK,eAAe;AAS/B,SAAO,EACL,MAAM;GACJ,OAPW,aAAa,SAAoB,SAAS,MAAM,MAAM,EAAE,UAAU,EAAE,EAAE;GAQjF,aALiB,aAAa,eAA0B,KAAK,mBAAmB,KAAK;GAMrF;GACA,OAAO;GACP;GACA;GACD,EACF;;CAGH,AAAQ,iBAAiB,SAA0C;EACjE,MAAME,SAAkC,EAAE;EAC1C,MAAM,QAAQ,QAAQ,MAAM,KAAK;AAEjC,OAAK,MAAM,QAAQ,OAAO;GACxB,MAAM,aAAa,KAAK,QAAQ,IAAI;AACpC,OAAI,aAAa,GAAG;IAClB,MAAM,MAAM,KAAK,MAAM,GAAG,WAAW,CAAC,MAAM;IAC5C,IAAIC,QAAmC,KAAK,MAAM,aAAa,EAAE,CAAC,MAAM;AAGxE,QAAI,UAAU,OAAQ,SAAQ;aACrB,UAAU,QAAS,SAAQ;aAC3B,kBAAkB,KAAK,MAAM,CAAE,SAAQ,OAAO,MAAM;aACpD,MAAM,WAAW,KAAI,IAAI,MAAM,SAAS,KAAI,CAAE,SAAQ,MAAM,MAAM,GAAG,GAAG;aACxE,MAAM,WAAW,IAAI,IAAI,MAAM,SAAS,IAAI,CAAE,SAAQ,MAAM,MAAM,GAAG,GAAG;AAEjF,WAAO,OAAO;;;AAIlB,SAAO;;CAGT,AAAQ,mBAAmB,MAAgC;AAEzD,OAAK,MAAM,QAAQ,KAAK,UAAU;AAChC,OAAI,KAAK,SAAS,UAAW;AAC7B,OAAI,KAAK,SAAS,YAChB,QAAOH,SAAc,KAAK;;;;;;;AAUlC,SAAgB,iBAAiB,UAAkC;CACjE,MAAM,OAAO,aAAa,SAAS;CACnC,MAAMI,QAAwB,EAAE;CAChC,IAAI,iBAAiB;AAErB,OAAM,OAAO,SAAS;AACpB,MAAI,KAAK,SAAS,UAChB,kBAAiBJ,SAAc,KAAgB;AAGjD,MAAI,KAAK,SAAS,WAIhB,OAHiB,MAGD,SAAS,aAAmB;AAC1C,SAAM,KAAK;IACT,KAAK,SAAS;IACd,MAAMA,SAAc,SAAS;IAC7B,OAAO,SAAS,SAAS;IACzB,SAAS,kBAAkB;IAC5B,CAAC;IACF;GAEJ;AAEF,QAAO;;;;;AAMT,SAAgB,gBAAgB,UAAkB,WAAW,GAAsB;AAGjF,QAFe,IAAI,gBAAgB,CACb,MAAM,SAAS,CACvB,KAAK,SAAS,QAAQ,MAAM,EAAE,SAAS,SAAS;;;;;;;;;;;;;;;;;;;;AC7KhE,IAAa,YAAb,MAAqE;CACnE,AAAS,OAAO;CAChB,AAAQ;CAER,YAAY,SAA4B;AACtC,OAAK,eAAe,SAAS,gBAAgB,EAAE;;CAGjD,SAAS,SAA0B;EACjC,MAAM,QAAQ,QAAQ,aAAa;AACnC,SAAO,MAAM,SAAS,OAAO,IAAI,MAAM,SAAS,QAAQ,IAAI,MAAM,SAAS,WAAW;;CAGxF,MAAM,SAAiB,KAAkD;EAGvE,MAAM,IAAI,QAAQ,KAAK,SAAS,EAAE,KAAK,MAAM,CAAC;AAG9C,MAAI,EAAE,OAAO,CAAC,SAAS,EACrB,QAAO,KAAK,UAAU,GAAG,IAAI;WACpB,EAAE,iBAAiB,CAAC,SAAS,EACtC,QAAO,KAAK,UAAU,GAAG,IAAI;MAE7B,QAAO,KAAK,UAAU,GAAG,IAAI;;CAIjC,AAAQ,UAAU,GAAuB,SAAsD;EAC7F,MAAM,UAAU,EAAE,UAAU;EAC5B,MAAM,WAAW,QAAQ,KAAK,SAAS,CAAC,MAAM;EAC9C,MAAM,cAAc,WAAW;EAE/B,MAAMK,QAAoB,EAAE,OAAO,CAChC,KAAK,GAAG,OAAO;GACd,MAAM,QAAQ,EAAE,GAAG;GACnB,MAAM,WAAW,MAAM,KAAK,OAAO,CAAC,MAAM;GAC1C,MAAM,OAAO,MAAM,KAAK,OAAO,CAAC,MAAM;GACtC,MAAM,UAAU,MAAM,KAAK,UAAU,CAAC,MAAM;GAG5C,MAAM,eAAe,KAAK,YAAY,UAAU,MAAM,YAAY;AAElE,UAAO;IACL,IAAI,QAAQ;IACZ,OAAO,MAAM,KAAK,QAAQ,CAAC,MAAM;IACjC,MAAM;IACN,aAAa,KAAK,UAAU,MAAM,KAAK,cAAc,CAAC;IACtD,SAAS,KAAK,oBAAoB,MAAM,KAAK,oBAAoB,CAAC;IAClE,QAAQ,MAAM,KAAK,SAAS,CAAC,MAAM,IAAI,MAAM,KAAK,eAAe,CAAC,MAAM,IAAI;IAC5E,aAAa,KAAK,UAAU,QAAQ;IACpC,gBAAgB,WAAW;IAC3B,YAAY,KAAK,gBACf,MACG,KAAK,WAAW,CAChB,KAAK,KAAG,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,CAC1B,KAAK,CACT;IACD,WAAW,KAAK,eAAe,MAAM,KAAK,YAAY,EAAE,YAAY;IACpE,cAAc,KAAK,oBAAoB,MAAM;IAC9C;IACD,CACD,KAAK;AAER,SAAO;GACL,MAAM;IACJ,QAAQ;IACR,OAAO,QAAQ,KAAK,UAAU,CAAC,MAAM;IACrC,aAAa,QAAQ,KAAK,gBAAgB,CAAC,MAAM,IAAI;IACrD,MAAM,KAAK,WAAW,UAAU,YAAY;IAC5C,UAAU,QAAQ,KAAK,aAAa,CAAC,MAAM,IAAI;IAC/C,eAAe,KAAK,UAAU,QAAQ,KAAK,kBAAkB,CAAC,MAAM,CAAC;IACrE,WAAW,QAAQ,KAAK,cAAc,CAAC,MAAM,IAAI;IACjD;IACA,cAAc,KAAK,oBAAoB,QAAQ;IAChD;GACD,MAAM;IACJ,WAAW,QAAQ,KAAK,cAAc,CAAC,MAAM,IAAI;IACjD,KAAK,KAAK,YAAY,QAAQ,KAAK,QAAQ,CAAC,MAAM,CAAC;IACnD,OAAO,KAAK,WAAW,QAAQ,KAAK,UAAU,EAAE,YAAY;IAC5D,YAAY,KAAK,gBACf,QACG,KAAK,aAAa,CAClB,KAAK,GAAG,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,CAC1B,KAAK,CACT;IACF;GACF;;CAGH,AAAQ,UAAU,GAAuB,SAAsD;EAC7F,MAAM,OAAO,EAAE,OAAO;EACtB,MAAM,WAAW,KAAK,KAAK,+CAA6C,CAAC,KAAK,OAAO,IAAI;EACzF,MAAM,WAAW,KAAK,KAAK,uBAAqB,CAAC,KAAK,OAAO;EAC7D,MAAM,cAAc,WAAW;EAE/B,MAAMA,QAAoB,EAAE,QAAQ,CACjC,KAAK,GAAG,OAAO;GACd,MAAM,SAAS,EAAE,GAAG;GACpB,MAAM,YAAY,OAAO,KAAK,2CAAyC,CAAC,KAAK,OAAO,IAAI;GACxF,MAAM,UAAU,OAAO,KAAK,KAAK,CAAC,MAAM;GACxC,MAAM,YAAY,OAAO,KAAK,YAAY,CAAC,MAAM;GACjD,MAAM,UAAU,OAAO,KAAK,UAAU,CAAC,MAAM;GAG7C,MAAM,eAAe,KAAK,YAAY,WAAW,SAAS,YAAY;AAEtE,UAAO;IACL,IAAI;IACJ,OAAO,OAAO,KAAK,QAAQ,CAAC,MAAM;IAClC,MAAM;IACN,aAAa,KAAK,UAAU,OAAO,KAAK,UAAU,CAAC;IACnD,SAAS,KAAK,UAAU,OAAO,KAAK,UAAU,CAAC;IAC/C,QAAQ,OAAO,KAAK,cAAc,CAAC,MAAM,IAAI;IAC7C,aAAa,KAAK,UAAU,UAAU;IACtC,gBAAgB,aAAa,WAAW;IACxC,WAAW,KAAK,UAAU,QAAQ;IAClC,YAAY,KAAK,gBACf,OACG,KAAK,WAAW,CAChB,KAAK,KAAG,MAAM,EAAE,EAAE,CAAC,KAAK,OAAO,CAAC,CAChC,KAAK,CACT;IACD,cAAc,KAAK,oBAAoB,OAAO;IAC/C;IACD,CACD,KAAK;AAER,SAAO;GACL,MAAM;IACJ,QAAQ;IACR,OAAO,KAAK,KAAK,UAAU,CAAC,MAAM;IAClC,aAAa,KAAK,KAAK,aAAa,CAAC,MAAM,IAAI;IAC/C,MAAM,KAAK,WAAW,UAAU,YAAY;IAC5C,MAAM,WAAW,KAAK,WAAW,UAAU,YAAY,GAAG;IAC1D,UAAU,KAAK,KAAK,WAAW,IAAI;IACnC,eAAe,KAAK,UAAU,KAAK,KAAK,YAAY,CAAC,MAAM,CAAC;IAC5D,WAAW,KAAK,KAAK,WAAW,CAAC,MAAM,IAAI;IAC3C;IACA,cAAc,KAAK,oBAAoB,KAAK;IAC7C;GACD,MAAM;IACJ,WAAW,KAAK,KAAK,cAAc,CAAC,MAAM,IAAI;IAC9C,OAAO,KAAK,eAAe,MAAM,YAAY;IAC7C,YAAY,KAAK,gBACf,KACG,KAAK,aAAa,CAClB,KAAK,GAAG,MAAM,EAAE,EAAE,CAAC,KAAK,OAAO,CAAC,CAChC,KAAK,CACT;IACF;GACF;;CAGH,AAAQ,UAAU,GAAuB,SAAsD;EAC7F,MAAM,UAAU,EAAE,UAAU;EAC5B,MAAM,WAAW,QAAQ,KAAK,OAAO,CAAC,MAAM;EAC5C,MAAM,cAAc,WAAW;EAG/B,MAAMA,QAAoB,EAAE,OAAO,CAChC,KAAK,GAAG,OAAO;GACd,MAAM,QAAQ,EAAE,GAAG;GACnB,MAAM,WAAW,MAAM,KAAK,OAAO,CAAC,MAAM;GAC1C,MAAM,WAAW,MAAM,KAAK,YAAY,IAAI;GAC5C,MAAM,SAAS,MAAM,KAAK,YAAY,CAAC,MAAM;GAG7C,MAAM,eAAe,KAAK,YAAY,UAAU,UAAU,YAAY;GAGtE,MAAM,aAAa,MAChB,KAAK,eAAe,CACpB,KAAK,KAAG,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,CAC1B,KAAK;AAER,UAAO;IACL,IAAI,YAAY;IAChB,OAAO,MAAM,KAAK,QAAQ,CAAC,MAAM;IACjC,MAAM;IACN,aAAa,KAAK,UAAU,MAAM,KAAK,cAAc,CAAC;IACtD,SAAS,KAAK,oBAAoB,MAAM,KAAK,oBAAoB,CAAC;IAClE,QAAQ,MAAM,KAAK,eAAe,CAAC,MAAM,IAAI;IAC7C,aAAa,KAAK,UAAU,OAAO;IACnC,gBAAgB,UAAU;IAC1B,YAAY,KAAK,gBAAgB,WAAW;IAC5C,cAAc,KAAK,oBAAoB,MAAM;IAC9C;IACD,CACD,KAAK;EAGR,MAAM,WAAW,EAAE,QAAQ;EAC3B,MAAM,WAAW,SAAS,KAAK,MAAM,CAAC,MAAM,IAAI,SAAS,KAAK,eAAe;AAE7E,SAAO;GACL,MAAM;IACJ,QAAQ;IACR,OAAO,QAAQ,KAAK,QAAQ,CAAC,MAAM;IACnC,aAAa,QAAQ,KAAK,cAAc,CAAC,MAAM,IAAI;IACnD,MAAM,KAAK,WAAW,UAAU,YAAY;IAC5C,UAAU,QAAQ,KAAK,gBAAgB,CAAC,MAAM,IAAI;IAClD,eAAe,KAAK,UAAU,QAAQ,KAAK,YAAY,CAAC,MAAM,CAAC;IAC/D,WAAW,QAAQ,KAAK,cAAc,CAAC,MAAM,IAAI;IACjD;IACA,cAAc,KAAK,oBAAoB,QAAQ;IAChD;GACD,MAAM;IACJ,WAAW,QAAQ,KAAK,yBAAyB,CAAC,KAAK,eAAe,IAAI;IAC1E,OAAO,WACH;KACE,KAAK,KAAK,WAAW,UAAU,YAAY;KAC3C,OAAO,SAAS,KAAK,QAAQ,CAAC,MAAM,IAAI;KACxC,MAAM,KAAK,WAAW,SAAS,KAAK,OAAO,CAAC,MAAM,EAAE,YAAY,IAAI;KACrE,GACD;IACJ,YAAY,KAAK,gBACf,QACG,KAAK,eAAe,CACpB,KAAK,GAAG,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,CAC1B,KAAK,CACT;IACF;GACF;;;;;CAMH,AAAQ,oBAAoB,KAAmE;AAC7F,MAAI,OAAO,KAAK,KAAK,aAAa,CAAC,WAAW,EAAG,QAAO;EAExD,MAAMC,SAAiC,EAAE;AACzC,OAAK,MAAM,CAAC,KAAK,aAAa,OAAO,QAAQ,KAAK,aAAa,EAAE;GAC/D,MAAM,QAAQ,IAAI,KAAK,SAAS,CAAC,MAAM,CAAC,MAAM;AAC9C,OAAI,MACF,QAAO,OAAO;;AAIlB,SAAO,OAAO,KAAK,OAAO,CAAC,SAAS,IAAI,SAAS;;;;;;CAOnD,AAAQ,UAAU,SAAqC;AACrD,MAAI,CAAC,SAAS,MAAM,CAAE,QAAO;AAE7B,MAAI;GACF,MAAM,OAAO,IAAI,KAAK,QAAQ;AAE9B,OAAI,OAAO,MAAM,KAAK,SAAS,CAAC,CAC9B;AAEF,UAAO,KAAK,aAAa;UACnB;AACN;;;;;;CAOJ,AAAQ,UAAU,KAAmD;AAEnE,SADa,IAAI,MAAM,CAAC,MAAM,IACf;;;;;CAMjB,AAAQ,oBAAoB,KAAmD;EAC7E,MAAM,MAAM,IAAI,MAAM,CAAC,MAAM;AAC7B,MAAI,CAAC,IAAK,QAAO;AACjB,SAAO,IAAI,QAAQ,YAAY,GAAG,CAAC,MAAM,IAAI;;;;;CAM/C,AAAQ,gBAAgB,YAA8C;AACpE,SAAO,WAAW,KAAK,MAAM,GAAG,MAAM,CAAC,CAAC,QAAQ,MAAmB,CAAC,CAAC,KAAK,EAAE,SAAS,EAAE;;;;;;CAOzF,AAAQ,WAAW,KAAa,MAAuB;AACrD,MAAI,CAAC,KAAK,MAAM,CAAE,QAAO;AAEzB,MAAI;GACF,MAAM,WAAW,OAAO,IAAI,IAAI,KAAK,KAAK,GAAG,IAAI,IAAI,IAAI;AACzD,UAAO,SAAS,aAAa,WAAW,SAAS,OAAO;UAClD;AACN,UAAO;;;;;;;CAQX,AAAQ,YAAY,aAAqB,YAAoB,MAAuB;AAElF,MAAI,aAAa,MAAM,CACrB,QAAO,KAAK,WAAW,aAAa,KAAK;AAI3C,MAAI,YAAY,MAAM,CACpB,KAAI;GACF,MAAM,aAAa,IAAI,IAAI,WAAW;AACtC,UAAO,WAAW,aAAa,WAAW,WAAW,OAAO;UACtD;AAEN,UAAO,KAAK,WAAW,YAAY,KAAK;;AAI5C,SAAO;;;;;CAMT,AAAQ,eACN,MACA,SAC2B;EAC3B,MAAM,MAAM,KAAK,KAAK,MAAM;AAC5B,MAAI,CAAC,IAAK,QAAO;AAEjB,SAAO;GACL,KAAK,KAAK,WAAW,KAAK,QAAQ;GAClC,MAAM,KAAK,KAAK,OAAO,IAAI;GAC3B,QAAQ,KAAK,YAAY,KAAK,KAAK,SAAS,CAAC;GAC9C;;;;;CAMH,AAAQ,WACN,MACA,SAC+B;EAC/B,MAAM,MAAM,KAAK,KAAK,MAAM,CAAC,MAAM;AACnC,MAAI,CAAC,IAAK,QAAO;AAEjB,SAAO;GACL,KAAK,KAAK,WAAW,KAAK,QAAQ;GAClC,OAAO,KAAK,KAAK,QAAQ,CAAC,MAAM,IAAI;GACpC,MAAM,KAAK,WAAW,KAAK,KAAK,OAAO,CAAC,MAAM,EAAE,QAAQ,IAAI;GAC7D;;;;;CAMH,AAAQ,eACN,OACA,SAC+B;EAC/B,MAAM,OAAO,MAAM,KAAK,SAAS,CAAC,MAAM;EACxC,MAAM,OAAO,MAAM,KAAK,SAAS,CAAC,MAAM;EACxC,MAAM,MAAM,QAAQ;AAEpB,MAAI,CAAC,IAAK,QAAO;AAEjB,SAAO,EACL,KAAK,KAAK,WAAW,KAAK,QAAQ,EACnC;;;;;CAMH,AAAQ,YAAY,OAA+C;AACjE,MAAI,CAAC,MAAO,QAAO;EACnB,MAAM,MAAM,SAAS,OAAO,GAAG;AAC/B,SAAO,OAAO,MAAM,IAAI,GAAG,SAAY"}