@ryangarber/markitdown-ts 0.1.4 → 0.1.5

package/README.md

@@ -10,6 +10,7 @@
 - Jupyter notebooks (`.ipynb`)
 - Word documents (`.docx`)
 - Excel workbooks (`.xlsx`)
+- PowerPoint presentations (`.pptx`)
 - PDF documents (`.pdf`)
 - ZIP archives (`.zip`), with nested files converted recursively
 - Wikipedia, YouTube metadata, and Bing result pages supplied as HTML

package/dist/cli.mjs

@@ -5,11 +5,12 @@ import TurndownService from "turndown";
 import { parseHTML } from "linkedom";
 import mammoth from "mammoth";
 import { extractText, getDocumentProxy, getMeta } from "unpdf";
-import ExcelJS from "exceljs";
-import { DOMParser } from "@xmldom/xmldom";
+import { pptxToHtml } from "@jvmr/pptx-to-html";
+import { DOMParser, Document, Element, Node } from "@xmldom/xmldom";
+import readExcelFile from "read-excel-file/universal";
 import { unzipSync } from "fflate";
 //#region package.json
-var version = "0.1.4";
+var version = "0.1.5";
 //#endregion
 //#region src/bytes.ts
 async function toUint8Array(source) {
@@ -220,6 +221,7 @@ const MIME_EXTENSIONS = {
 	"application/zip": ".zip",
 	"application/rss+xml": ".rss",
 	"application/x-pdf": ".pdf",
+	"application/vnd.openxmlformats-officedocument.presentationml.presentation": ".pptx",
 	"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet": ".xlsx",
 	"application/vnd.openxmlformats-officedocument.wordprocessingml.document": ".docx",
 	"application/xml": ".xml",
@@ -265,6 +267,85 @@ var PlainTextConverter = class {
 	}
 };
 //#endregion
+//#region src/converters/pptx.ts
+var PptxConverter = class extends HtmlConverter {
+	async convert(source, options) {
+		if (options.file_extension?.toLowerCase() !== ".pptx") return null;
+		const slides = await pptxToHtml(toArrayBuffer(source), { domParserFactory: createPptxDomParser });
+		const sections = [];
+		for (const [index, html] of slides.entries()) {
+			const result = await this.convertHtml(`<html><body>${html}</body></html>`);
+			sections.push(`<!-- Slide number: ${index + 1} -->\n${result?.markdown.trim() ?? ""}`);
+		}
+		const markdown = sections.join("\n\n");
+		return {
+			title: null,
+			markdown,
+			text_content: markdown
+		};
+	}
+};
+function createPptxDomParser() {
+	if (typeof globalThis.DOMParser === "function") return new globalThis.DOMParser();
+	patchXmlDom();
+	return new DOMParser();
+}
+let xmlDomPatched = false;
+function patchXmlDom() {
+	if (xmlDomPatched) return;
+	xmlDomPatched = true;
+	defineMethod(Element.prototype, "querySelector", querySelector);
+	defineMethod(Element.prototype, "querySelectorAll", querySelectorAll);
+	defineMethod(Document.prototype, "querySelector", querySelector);
+	defineMethod(Document.prototype, "querySelectorAll", querySelectorAll);
+	defineGetter(Element.prototype, "children", function() {
+		return [...this.childNodes].filter((node) => node.nodeType === 1);
+	});
+	defineGetter(Node.prototype, "parentElement", function() {
+		const parent = this.parentNode;
+		return parent?.nodeType === 1 ? parent : null;
+	});
+}
+function querySelector(selector) {
+	for (const group of selector.split(",")) {
+		const match = queryByPath(this, group.trim());
+		if (match) return match;
+	}
+	return null;
+}
+function querySelectorAll() {
+	return [];
+}
+function queryByPath(root, selector) {
+	const tokens = selector.split(/\s+/).filter(Boolean);
+	let current = [root];
+	for (const token of tokens) {
+		const localName = selectorLocalName(token);
+		const next = [];
+		for (const node of current) next.push(...Array.from(node.getElementsByTagNameNS("*", localName)));
+		current = next;
+	}
+	return current[0] ?? null;
+}
+function selectorLocalName(token) {
+	if (token.includes("|")) return token.slice(token.lastIndexOf("|") + 1);
+	return token.replace(/\\:/g, ":").split(":").pop() ?? token;
+}
+function defineMethod(target, name, value) {
+	Object.defineProperty(target, name, {
+		configurable: true,
+		writable: true,
+		value
+	});
+}
+function defineGetter(target, name, get) {
+	if (Object.getOwnPropertyDescriptor(target, name)) return;
+	Object.defineProperty(target, name, {
+		configurable: true,
+		get
+	});
+}
+//#endregion
 //#region src/converters/wikipedia.ts
 const WIKIPEDIA_URL = /^https?:\/\/[a-z]{2,3}\.wikipedia\.org\//i;
 var WikipediaConverter = class {
@@ -287,41 +368,38 @@ var WikipediaConverter = class {
 };
 //#endregion
 //#region src/converters/xlsx.ts
-const { Workbook } = ExcelJS;
 var XlsxConverter = class extends HtmlConverter {
 	async convert(source, options) {
 		if (options.file_extension?.toLowerCase() !== ".xlsx") return null;
-		const workbook = new Workbook();
-		await workbook.xlsx.load(toArrayBuffer(source));
+		const sheets = await readExcelFile(toArrayBuffer(source));
 		const sections = [];
-		for (const worksheet of workbook.worksheets) {
-			const html = worksheetToHtml(worksheet);
+		for (const { sheet, data } of sheets) {
+			const html = sheetToHtml(data);
 			if (!html) continue;
 			const result = await this.convertHtml(html);
-			sections.push(`## ${worksheet.name}\n${result?.markdown.trim() ?? ""}`);
+			sections.push(`## ${sheet}\n${result?.markdown.trim() ?? ""}`);
 		}
 		const markdown = sections.join("\n\n");
 		return {
-			title: workbook.title || "Untitled",
+			title: null,
 			markdown,
 			text_content: markdown
 		};
 	}
 };
-function worksheetToHtml(worksheet) {
-	const columnCount = worksheet.columnCount;
-	if (!columnCount || !worksheet.actualRowCount) return "";
-	const rows = [];
-	worksheet.eachRow({ includeEmpty: true }, (row) => {
+function sheetToHtml(rows) {
+	const columnCount = rows.reduce((max, row) => Math.max(max, row.length), 0);
+	if (!columnCount || rows.length === 0) return "";
+	return `<html><body><table>${rows.map((row) => {
 		const cells = [];
-		for (let column = 1; column <= columnCount; column++) cells.push(`<td>${escapeHtml(cellText(row.getCell(column)))}</td>`);
-		rows.push(`<tr>${cells.join("")}</tr>`);
-	});
-	return `<html><body><table>${rows.join("")}</table></body></html>`;
+		for (let column = 0; column < columnCount; column++) cells.push(`<td>${escapeHtml(cellText(row[column]))}</td>`);
+		return `<tr>${cells.join("")}</tr>`;
+	}).join("")}</table></body></html>`;
 }
-function cellText(cell) {
-	if (cell.isMerged && cell.master !== cell) return "";
-	return cell.text;
+function cellText(value) {
+	if (value == null) return "";
+	if (value instanceof Date) return value.toISOString();
+	return String(value);
 }
 function escapeHtml(value) {
 	return value.replace(/&/g, "&amp;").replace(/</g, "&lt;").replace(/>/g, "&gt;").replace(/"/g, "&quot;");
@@ -520,6 +598,7 @@ var MarkItDown = class {
 		this.registerConverter(new BingSerpConverter());
 		this.registerConverter(new DocxConverter());
 		this.registerConverter(new XlsxConverter());
+		this.registerConverter(new PptxConverter());
 		this.registerConverter(new PdfConverter());
 		this.registerConverter(new IpynbConverter());
 	}