@bendyline/squisq-formats 0.1.0

package/src/html/index.ts

@@ -0,0 +1,168 @@
+/**
+ * HTML Export Module — @bendyline/squisq-formats/html
+ *
+ * Exports squisq documents as self-contained HTML files or ZIP archives.
+ *
+ * Two export modes:
+ *
+ * 1. **Single HTML** (`docToHtml`) — Everything embedded in one file:
+ *    - SquisqPlayer IIFE bundle inlined in `<script>`
+ *    - Images base64-encoded as data URIs
+ *    - Timer-based playback (no audio — too large for inline)
+ *
+ * 2. **ZIP Archive** (`docToHtmlZip`) — Multi-file package:
+ *    - `index.html` referencing external `squisq-player.js`
+ *    - `squisq-player.js` — the IIFE bundle
+ *    - `images/` — extracted image files
+ *    - `audio/` — optional audio segment files (enables full playback)
+ *
+ * @example
+ * ```ts
+ * import { PLAYER_BUNDLE } from '@bendyline/squisq-react/standalone-source';
+ * import { docToHtml, docToHtmlZip } from '@bendyline/squisq-formats/html';
+ *
+ * // Single HTML file
+ * const html = docToHtml(doc, { playerScript: PLAYER_BUNDLE, images });
+ *
+ * // ZIP archive
+ * const zipBlob = await docToHtmlZip(doc, { playerScript: PLAYER_BUNDLE, images, audio });
+ * ```
+ */
+
+import type { Doc } from '@bendyline/squisq/schemas';
+import JSZip from 'jszip';
+import {
+  generateInlineHtml,
+  generateExternalHtml,
+  type HtmlExportOptions,
+} from './htmlTemplate.js';
+import { extractFilename } from './imageUtils.js';
+
+// ── Public Types ───────────────────────────────────────────────────
+
+export type { HtmlExportOptions };
+
+export interface HtmlZipExportOptions extends HtmlExportOptions {
+  /**
+   * Map of audio segment identifiers to binary audio data.
+   * Keys should match the audio segment `name` or `url` fields in the Doc.
+   * When provided, audio files are included in the ZIP and full playback is enabled.
+   */
+  audio?: Map<string, ArrayBuffer>;
+}
+
+// ── Single HTML Export ─────────────────────────────────────────────
+
+/**
+ * Export a Doc as a single, self-contained HTML file.
+ *
+ * The player JS is embedded inline, images are base64-encoded as data URIs,
+ * and playback uses a timer (no audio) for compact file size.
+ *
+ * @param doc - The Doc to export
+ * @param options - Export options (must include `playerScript`)
+ * @returns Complete HTML document as a string
+ *
+ * @example
+ * ```ts
+ * const html = docToHtml(myDoc, {
+ *   playerScript: PLAYER_BUNDLE,
+ *   images: imageMap,
+ *   title: 'My Document',
+ *   mode: 'slideshow',
+ * });
+ * const blob = new Blob([html], { type: 'text/html' });
+ * ```
+ */
+export function docToHtml(doc: Doc, options: HtmlExportOptions): string {
+  return generateInlineHtml(doc, options);
+}
+
+// ── ZIP Archive Export ─────────────────────────────────────────────
+
+/**
+ * Export a Doc as a ZIP archive containing HTML, JS, images, and optionally audio.
+ *
+ * The archive structure:
+ * ```
+ * document.zip
+ * ├── index.html           # HTML page referencing squisq-player.js
+ * ├── squisq-player.js     # Standalone IIFE bundle
+ * ├── images/              # Extracted image files
+ * │   ├── hero.jpg
+ * │   └── ...
+ * └── audio/               # Optional audio segment files
+ *     ├── intro.mp3
+ *     └── ...
+ * ```
+ *
+ * @param doc - The Doc to export
+ * @param options - Export options (must include `playerScript`)
+ * @returns A Promise resolving to a ZIP Blob
+ *
+ * @example
+ * ```ts
+ * const blob = await docToHtmlZip(myDoc, {
+ *   playerScript: PLAYER_BUNDLE,
+ *   images: imageMap,
+ *   audio: audioMap,
+ * });
+ * // Trigger browser download
+ * const url = URL.createObjectURL(blob);
+ * ```
+ */
+export async function docToHtmlZip(doc: Doc, options: HtmlZipExportOptions): Promise<Blob> {
+  const { playerScript, images, audio, mode = 'slideshow', title, autoPlay } = options;
+
+  const zip = new JSZip();
+
+  // 1. Add player JS as a separate file
+  zip.file('squisq-player.js', playerScript);
+
+  // 2. Add images to images/ folder and build path mapping
+  const imagePathMap: Record<string, string> = {};
+  if (images) {
+    for (const [originalPath, buffer] of images.entries()) {
+      const filename = extractFilename(originalPath);
+      const zipPath = `images/${filename}`;
+      zip.file(zipPath, buffer);
+      imagePathMap[originalPath] = zipPath;
+    }
+  }
+
+  // 3. Add audio to audio/ folder and build path mapping
+  const audioPathMap: Record<string, string> = {};
+  if (audio) {
+    for (const [segmentKey, buffer] of audio.entries()) {
+      const filename = extractFilename(segmentKey);
+      // Ensure .mp3 extension
+      const finalName = filename.includes('.') ? filename : `${filename}.mp3`;
+      const zipPath = `audio/${finalName}`;
+      zip.file(zipPath, buffer);
+      audioPathMap[segmentKey] = zipPath;
+    }
+  }
+
+  // 4. Generate HTML that references external files
+  const html = generateExternalHtml(doc, {
+    playerScriptPath: 'squisq-player.js',
+    imagePathMap: Object.keys(imagePathMap).length > 0 ? imagePathMap : undefined,
+    audioPathMap: Object.keys(audioPathMap).length > 0 ? audioPathMap : undefined,
+    mode,
+    title,
+    autoPlay,
+  });
+  zip.file('index.html', html);
+
+  // 5. Generate ZIP blob
+  return zip.generateAsync({
+    type: 'blob',
+    compression: 'DEFLATE',
+    compressionOptions: { level: 6 },
+  });
+}
+
+// ── Re-exports ─────────────────────────────────────────────────────
+
+export { collectImagePaths } from './htmlTemplate.js';
+export { inferMimeType, arrayBufferToBase64DataUrl, extractFilename } from './imageUtils.js';

package/src/index.ts

@@ -0,0 +1,51 @@
+/**
+ * @bendyline/squisq-formats
+ *
+ * Format converters for squisq documents. Converts between squisq's
+ * MarkdownDocument / Doc and various file formats via Office Open XML.
+ *
+ * Supported formats:
+ * - **DOCX** — Microsoft Word (import + export) ✅
+ * - **PDF**  — Portable Document Format (import + export) ✅
+ * - **PPTX** — Microsoft PowerPoint (planned)
+ * - **XLSX** — Microsoft Excel (planned)
+ *
+ * All converters run in the browser — no server or native binaries required.
+ * The shared `ooxml/` subpath export provides reusable OOXML infrastructure.
+ *
+ * @example
+ * ```ts
+ * // Import from root
+ * import { markdownDocToDocx, docxToMarkdownDoc } from '@bendyline/squisq-formats';
+ *
+ * // Or import from subpath
+ * import { markdownDocToDocx } from '@bendyline/squisq-formats/docx';
+ * import { createPackage } from '@bendyline/squisq-formats/ooxml';
+ * ```
+ */
+
+// DOCX (fully implemented)
+export { markdownDocToDocx, docToDocx, docxToMarkdownDoc, docxToDoc } from './docx/index.js';
+export type { DocxExportOptions, DocxImportOptions } from './docx/index.js';
+
+// PPTX (stub)
+export { markdownDocToPptx, docToPptx, pptxToMarkdownDoc, pptxToDoc } from './pptx/index.js';
+export type { PptxExportOptions, PptxImportOptions } from './pptx/index.js';
+
+// XLSX (stub)
+export { markdownDocToXlsx, docToXlsx, xlsxToMarkdownDoc, xlsxToDoc } from './xlsx/index.js';
+export type { XlsxExportOptions, XlsxImportOptions } from './xlsx/index.js';
+
+// PDF (fully implemented)
+export {
+  markdownDocToPdf,
+  docToPdf,
+  pdfToMarkdownDoc,
+  pdfToDoc,
+  configurePdfWorker,
+} from './pdf/index.js';
+export type { PdfExportOptions, PdfImportOptions } from './pdf/index.js';
+
+// HTML (fully implemented)
+export { docToHtml, docToHtmlZip, collectImagePaths } from './html/index.js';
+export type { HtmlExportOptions, HtmlZipExportOptions } from './html/index.js';

package/src/ooxml/index.ts

@@ -0,0 +1,87 @@
+/**
+ * @bendyline/squisq-formats OOXML Module
+ *
+ * Shared infrastructure for reading and writing Office Open XML packages
+ * (DOCX, PPTX, XLSX). Provides ZIP archive handling, relationship management,
+ * content type assembly, XML utilities, and core property parsing.
+ *
+ * This module is consumed by the format-specific modules (docx/, pptx/, xlsx/)
+ * and can also be imported directly by advanced consumers.
+ *
+ * @example
+ * ```ts
+ * import { createPackage, openPackage } from '@bendyline/squisq-formats/ooxml';
+ * ```
+ */
+
+// Types
+export type {
+  OoxmlPackage,
+  ContentTypeMap,
+  Relationship,
+  CoreProperties,
+  PackagePart,
+  PendingRelationship,
+} from './types.js';
+
+// Reader
+export {
+  openPackage,
+  getPartRelationships,
+  getPartXml,
+  getPartBinary,
+  getCoreProperties,
+} from './reader.js';
+
+// Writer
+export type { OoxmlPackageBuilder } from './writer.js';
+export { createPackage } from './writer.js';
+
+// XML Utilities
+export {
+  xmlDeclaration,
+  escapeXml,
+  attrString,
+  selfClosingElement,
+  xmlElement,
+  textElement,
+} from './xmlUtils.js';
+
+// Namespace constants
+export {
+  NS_RELATIONSHIPS,
+  NS_CONTENT_TYPES,
+  REL_OFFICE_DOCUMENT,
+  REL_CORE_PROPERTIES,
+  REL_EXTENDED_PROPERTIES,
+  REL_STYLES,
+  REL_NUMBERING,
+  REL_FONT_TABLE,
+  REL_SETTINGS,
+  REL_HYPERLINK,
+  REL_IMAGE,
+  REL_FOOTNOTES,
+  REL_THEME,
+  NS_WML,
+  NS_PML,
+  NS_SML,
+  NS_DRAWINGML,
+  NS_WP_DRAWING,
+  NS_PICTURE,
+  NS_DC,
+  NS_DCTERMS,
+  NS_CORE_PROPERTIES,
+  NS_XSI,
+  NS_MC,
+  NS_R,
+  CONTENT_TYPE_RELATIONSHIPS,
+  CONTENT_TYPE_CORE_PROPERTIES,
+  CONTENT_TYPE_DOCX_DOCUMENT,
+  CONTENT_TYPE_DOCX_STYLES,
+  CONTENT_TYPE_DOCX_NUMBERING,
+  CONTENT_TYPE_DOCX_SETTINGS,
+  CONTENT_TYPE_DOCX_FONT_TABLE,
+  CONTENT_TYPE_DOCX_FOOTNOTES,
+  CONTENT_TYPE_PPTX_PRESENTATION,
+  CONTENT_TYPE_XLSX_WORKBOOK,
+} from './namespaces.js';

package/src/ooxml/namespaces.ts

@@ -0,0 +1,160 @@
+/**
+ * OOXML Namespace Constants
+ *
+ * All Office Open XML namespace URIs used across DOCX, PPTX, and XLSX.
+ * Organized by category for easy reference.
+ */
+
+// ============================================
+// Package-level namespaces (shared by all formats)
+// ============================================
+
+/** Relationships namespace (used in _rels/*.rels files) */
+export const NS_RELATIONSHIPS = 'http://schemas.openxmlformats.org/package/2006/relationships';
+
+/** Content Types namespace ([Content_Types].xml) */
+export const NS_CONTENT_TYPES = 'http://schemas.openxmlformats.org/package/2006/content-types';
+
+// ============================================
+// Relationship type URIs
+// ============================================
+
+/** Relationship type: Office document (main part) */
+export const REL_OFFICE_DOCUMENT =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument';
+
+/** Relationship type: Core properties */
+export const REL_CORE_PROPERTIES =
+  'http://schemas.openxmlformats.org/package/2006/relationships/metadata/core-properties';
+
+/** Relationship type: Extended properties (app.xml) */
+export const REL_EXTENDED_PROPERTIES =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/extended-properties';
+
+/** Relationship type: Styles */
+export const REL_STYLES =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles';
+
+/** Relationship type: Numbering */
+export const REL_NUMBERING =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/numbering';
+
+/** Relationship type: Font table */
+export const REL_FONT_TABLE =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/fontTable';
+
+/** Relationship type: Settings */
+export const REL_SETTINGS =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/settings';
+
+/** Relationship type: Hyperlink */
+export const REL_HYPERLINK =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/hyperlink';
+
+/** Relationship type: Image */
+export const REL_IMAGE =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/image';
+
+/** Relationship type: Footnotes */
+export const REL_FOOTNOTES =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/footnotes';
+
+/** Relationship type: Theme */
+export const REL_THEME =
+  'http://schemas.openxmlformats.org/officeDocument/2006/relationships/theme';
+
+// ============================================
+// WordprocessingML (DOCX)
+// ============================================
+
+/** WordprocessingML main namespace (w:) */
+export const NS_WML = 'http://schemas.openxmlformats.org/wordprocessingml/2006/main';
+
+// ============================================
+// PresentationML (PPTX)
+// ============================================
+
+/** PresentationML main namespace (p:) */
+export const NS_PML = 'http://schemas.openxmlformats.org/presentationml/2006/main';
+
+// ============================================
+// SpreadsheetML (XLSX)
+// ============================================
+
+/** SpreadsheetML main namespace */
+export const NS_SML = 'http://schemas.openxmlformats.org/spreadsheetml/2006/main';
+
+// ============================================
+// DrawingML (shared across formats)
+// ============================================
+
+/** DrawingML main namespace (a:) */
+export const NS_DRAWINGML = 'http://schemas.openxmlformats.org/drawingml/2006/main';
+
+/** DrawingML WordprocessingML drawing namespace (wp:) */
+export const NS_WP_DRAWING =
+  'http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing';
+
+/** DrawingML picture namespace (pic:) */
+export const NS_PICTURE = 'http://schemas.openxmlformats.org/drawingml/2006/picture';
+
+// ============================================
+// Dublin Core / Core Properties
+// ============================================
+
+/** Dublin Core elements namespace */
+export const NS_DC = 'http://purl.org/dc/elements/1.1/';
+
+/** Dublin Core terms namespace */
+export const NS_DCTERMS = 'http://purl.org/dc/terms/';
+
+/** Core properties namespace */
+export const NS_CORE_PROPERTIES =
+  'http://schemas.openxmlformats.org/package/2006/metadata/core-properties';
+
+/** XML Schema Instance namespace */
+export const NS_XSI = 'http://www.w3.org/2001/XMLSchema-instance';
+
+// ============================================
+// Markup Compatibility
+// ============================================
+
+/** Markup Compatibility namespace (mc:) */
+export const NS_MC = 'http://schemas.openxmlformats.org/markup-compatibility/2006';
+
+/** Office relationships namespace (r:) */
+export const NS_R = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships';
+
+// ============================================
+// Content type strings
+// ============================================
+
+export const CONTENT_TYPE_RELATIONSHIPS =
+  'application/vnd.openxmlformats-package.relationships+xml';
+
+export const CONTENT_TYPE_CORE_PROPERTIES =
+  'application/vnd.openxmlformats-package.core-properties+xml';
+
+export const CONTENT_TYPE_DOCX_DOCUMENT =
+  'application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml';
+
+export const CONTENT_TYPE_DOCX_STYLES =
+  'application/vnd.openxmlformats-officedocument.wordprocessingml.styles+xml';
+
+export const CONTENT_TYPE_DOCX_NUMBERING =
+  'application/vnd.openxmlformats-officedocument.wordprocessingml.numbering+xml';
+
+export const CONTENT_TYPE_DOCX_SETTINGS =
+  'application/vnd.openxmlformats-officedocument.wordprocessingml.settings+xml';
+
+export const CONTENT_TYPE_DOCX_FONT_TABLE =
+  'application/vnd.openxmlformats-officedocument.wordprocessingml.fontTable+xml';
+
+export const CONTENT_TYPE_DOCX_FOOTNOTES =
+  'application/vnd.openxmlformats-officedocument.wordprocessingml.footnotes+xml';
+
+export const CONTENT_TYPE_PPTX_PRESENTATION =
+  'application/vnd.openxmlformats-officedocument.presentationml.presentation.main+xml';
+
+export const CONTENT_TYPE_XLSX_WORKBOOK =
+  'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet.main+xml';

package/src/ooxml/reader.ts

@@ -0,0 +1,218 @@
+/**
+ * OOXML Package Reader
+ *
+ * Opens OOXML archives (.docx, .pptx, .xlsx) and parses their
+ * structural metadata: [Content_Types].xml, relationships, and
+ * core properties.
+ *
+ * Uses JSZip to unzip and the browser's DOMParser to parse XML.
+ */
+
+import JSZip from 'jszip';
+import type { OoxmlPackage, ContentTypeMap, Relationship, CoreProperties } from './types.js';
+import { NS_RELATIONSHIPS, NS_DC, NS_DCTERMS, NS_CORE_PROPERTIES } from './namespaces.js';
+
+// ============================================
+// Package Opening
+// ============================================
+
+/**
+ * Open an OOXML package from raw data.
+ *
+ * Parses the ZIP archive, [Content_Types].xml, and root relationships.
+ *
+ * @param data - The raw .docx/.pptx/.xlsx file as ArrayBuffer or Blob
+ * @returns A parsed OoxmlPackage
+ */
+export async function openPackage(data: ArrayBuffer | Blob): Promise<OoxmlPackage> {
+  const zip = await JSZip.loadAsync(data);
+  const contentTypes = await parseContentTypes(zip);
+  const rootRelationships = await parseRelationships(zip, '');
+
+  return { zip, contentTypes, rootRelationships };
+}
+
+// ============================================
+// Content Types
+// ============================================
+
+/**
+ * Parse [Content_Types].xml from the archive.
+ */
+async function parseContentTypes(zip: JSZip): Promise<ContentTypeMap> {
+  const overrides = new Map<string, string>();
+  const defaults = new Map<string, string>();
+
+  const file = zip.file('[Content_Types].xml');
+  if (!file) return { overrides, defaults };
+
+  const text = await file.async('text');
+  const doc = new DOMParser().parseFromString(text, 'application/xml');
+
+  // Parse <Default Extension="rels" ContentType="..." />
+  const defaultEls = doc.getElementsByTagName('Default');
+  for (let i = 0; i < defaultEls.length; i++) {
+    const el = defaultEls[i];
+    const ext = el.getAttribute('Extension');
+    const ct = el.getAttribute('ContentType');
+    if (ext && ct) defaults.set(ext, ct);
+  }
+
+  // Parse <Override PartName="/word/document.xml" ContentType="..." />
+  const overrideEls = doc.getElementsByTagName('Override');
+  for (let i = 0; i < overrideEls.length; i++) {
+    const el = overrideEls[i];
+    const partName = el.getAttribute('PartName');
+    const ct = el.getAttribute('ContentType');
+    if (partName && ct) {
+      // Normalize: strip leading slash
+      overrides.set(partName.replace(/^\//, ''), ct);
+    }
+  }
+
+  return { overrides, defaults };
+}
+
+// ============================================
+// Relationships
+// ============================================
+
+/**
+ * Parse relationships for a specific part.
+ *
+ * @param pkg - The OOXML package (or the zip directly)
+ * @param partPath - The part path (e.g., "word/document.xml").
+ *   Use "" for root-level relationships (_rels/.rels).
+ * @returns Array of relationship entries
+ */
+export async function getPartRelationships(
+  pkg: OoxmlPackage,
+  partPath: string,
+): Promise<Relationship[]> {
+  return parseRelationships(pkg.zip, partPath);
+}
+
+/**
+ * Parse a _rels/*.rels file from the ZIP.
+ *
+ * For root relationships, relsPath = "_rels/.rels".
+ * For part relationships, relsPath = "<dir>/_rels/<filename>.rels".
+ */
+async function parseRelationships(zip: JSZip, partPath: string): Promise<Relationship[]> {
+  const relsPath = partPath === '' ? '_rels/.rels' : buildRelsPath(partPath);
+
+  const file = zip.file(relsPath);
+  if (!file) return [];
+
+  const text = await file.async('text');
+  const doc = new DOMParser().parseFromString(text, 'application/xml');
+  const result: Relationship[] = [];
+
+  const els = doc.getElementsByTagNameNS(NS_RELATIONSHIPS, 'Relationship');
+  // Fallback in case namespace isn't used (some generators omit it)
+  const fallbackEls = els.length > 0 ? els : doc.getElementsByTagName('Relationship');
+
+  for (let i = 0; i < fallbackEls.length; i++) {
+    const el = fallbackEls[i];
+    const id = el.getAttribute('Id');
+    const type = el.getAttribute('Type');
+    const target = el.getAttribute('Target');
+    if (id && type && target) {
+      const targetMode = el.getAttribute('TargetMode') as 'Internal' | 'External' | null;
+      result.push({
+        id,
+        type,
+        target,
+        ...(targetMode ? { targetMode } : {}),
+      });
+    }
+  }
+
+  return result;
+}
+
+/**
+ * Build the _rels path for a given part path.
+ *
+ * "word/document.xml" → "word/_rels/document.xml.rels"
+ * "xl/workbook.xml"   → "xl/_rels/workbook.xml.rels"
+ */
+function buildRelsPath(partPath: string): string {
+  const lastSlash = partPath.lastIndexOf('/');
+  if (lastSlash === -1) {
+    return `_rels/${partPath}.rels`;
+  }
+  const dir = partPath.substring(0, lastSlash);
+  const file = partPath.substring(lastSlash + 1);
+  return `${dir}/_rels/${file}.rels`;
+}
+
+// ============================================
+// Part Access
+// ============================================
+
+/**
+ * Extract an XML part from the package and parse it as a DOM Document.
+ *
+ * @param pkg - The OOXML package
+ * @param partPath - Path within the archive (e.g., "word/document.xml")
+ * @returns Parsed XML Document, or null if the part doesn't exist
+ */
+export async function getPartXml(pkg: OoxmlPackage, partPath: string): Promise<Document | null> {
+  const file = pkg.zip.file(partPath);
+  if (!file) return null;
+
+  const text = await file.async('text');
+  return new DOMParser().parseFromString(text, 'application/xml');
+}
+
+/**
+ * Extract a binary part from the package (e.g., an image from word/media/).
+ *
+ * @param pkg - The OOXML package
+ * @param partPath - Path within the archive
+ * @returns The binary content, or null if the part doesn't exist
+ */
+export async function getPartBinary(
+  pkg: OoxmlPackage,
+  partPath: string,
+): Promise<ArrayBuffer | null> {
+  const file = pkg.zip.file(partPath);
+  if (!file) return null;
+  return file.async('arraybuffer');
+}
+
+// ============================================
+// Core Properties
+// ============================================
+
+/**
+ * Parse core document properties from docProps/core.xml.
+ *
+ * @param pkg - The OOXML package
+ * @returns Parsed core properties (all fields optional)
+ */
+export async function getCoreProperties(pkg: OoxmlPackage): Promise<CoreProperties> {
+  const doc = await getPartXml(pkg, 'docProps/core.xml');
+  if (!doc) return {};
+
+  function getText(ns: string, localName: string): string | undefined {
+    const els = doc!.getElementsByTagNameNS(ns, localName);
+    if (els.length > 0 && els[0].textContent) {
+      return els[0].textContent;
+    }
+    return undefined;
+  }
+
+  return {
+    title: getText(NS_DC, 'title'),
+    subject: getText(NS_DC, 'subject'),
+    creator: getText(NS_DC, 'creator'),
+    description: getText(NS_DC, 'description'),
+    keywords: getText(NS_CORE_PROPERTIES, 'keywords'),
+    lastModifiedBy: getText(NS_CORE_PROPERTIES, 'lastModifiedBy'),
+    revision: getText(NS_CORE_PROPERTIES, 'revision'),
+    created: getText(NS_DCTERMS, 'created'),
+    modified: getText(NS_DCTERMS, 'modified'),
+  };
+}