@adeu/core 1.7.1 → 1.7.4

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@adeu/core",
-  "version": "1.7.1",
+  "version": "1.7.4",
   "description": "",
   "main": "./dist/index.js",
   "types": "./dist/index.d.ts",
@@ -20,6 +20,9 @@
     "url": "https://github.com/dealfluence/adeu.git",
     "directory": "node/packages/core"
   },
+  "engines": {
+    "node": ">=22.0.0"
+  },
   "publishConfig": {
     "access": "public"
   },
@@ -29,7 +32,7 @@
   "dependencies": {
     "@xmldom/xmldom": "^0.9.10",
     "diff-match-patch": "^1.0.5",
-    "jszip": "^3.10.1",
+    "fflate": "^0.8.2",
     "xpath": "^0.0.34"
   },
   "devDependencies": {

package/src/docx/bridge.ts

@@ -1,4 +1,5 @@
-import JSZip from "jszip";
+// FILE: node/packages/core/src/docx/bridge.ts
+import { unzipSync, zipSync, strFromU8, strToU8 } from "fflate";
 import {
   parseXml,
   findChild,
@@ -59,10 +60,10 @@ export class DocxPackage {
   public parts: Part[] = [];
   public mainDocumentPart!: Part;
 
-  constructor(public zip: JSZip) {}
+  constructor(public unzipped: Record<string, Uint8Array>) {}
 
   public getPartByPath(path: string): Part | undefined {
-    // Strip leading slash for jszip compat
+    // Strip leading slash for zip compat
     const searchPath = path.startsWith("/") ? path.substring(1) : path;
     return this.parts.find(
       (p) => p.partname === searchPath || p.partname === "/" + searchPath,
@@ -151,14 +152,15 @@ export class DocumentObject {
   public static async load(
     buffer: Buffer | ArrayBuffer,
   ): Promise<DocumentObject> {
-    const zip = await JSZip.loadAsync(buffer);
-    const pkg = new DocxPackage(zip);
+    const u8 = buffer instanceof Uint8Array ? buffer : new Uint8Array(buffer);
+    const unzipped = unzipSync(u8);
+    const pkg = new DocxPackage(unzipped);
 
     // 1. Load Content Types
-    const ctFile = zip.file("[Content_Types].xml");
+    const ctFile = unzipped["[Content_Types].xml"];
     let contentTypes: Record<string, string> = {};
     if (ctFile) {
-      const ctXml = parseXml(await ctFile.async("text"));
+      const ctXml = parseXml(strFromU8(ctFile));
       const overrides = findAllDescendants(ctXml.documentElement, "Override");
       for (const override of overrides) {
         contentTypes[override.getAttribute("PartName") || ""] =
@@ -167,9 +169,9 @@ export class DocumentObject {
     }
 
     // 2. Pre-load all XML parts to allow synchronous traversal later
-    for (const [path, file] of Object.entries(zip.files)) {
-      if (!file.dir && (path.endsWith(".xml") || path.endsWith(".rels"))) {
-        const text = await file.async("text");
+    for (const [path, fileData] of Object.entries(unzipped)) {
+      if (path.endsWith(".xml") || path.endsWith(".rels")) {
+        const text = strFromU8(fileData);
         const doc = parseXml(text);
         const cType = contentTypes["/" + path] || "application/xml";
         const part = new Part("/" + path, text, doc.documentElement, cType);
@@ -236,8 +238,8 @@ export class DocumentObject {
         xmlStr =
           '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n' + xmlStr;
       }
-      this.pkg.zip.file(part.partname.substring(1), xmlStr); // Strip leading slash for JSZip
+      this.pkg.unzipped[part.partname.substring(1)] = strToU8(xmlStr); // Strip leading slash
     }
-    return this.pkg.zip.generateAsync({ type: "nodebuffer" });
+    return Buffer.from(zipSync(this.pkg.unzipped));
   }
 }

package/src/sanitize/sanitize.test.ts

@@ -1,60 +1,63 @@
-import { describe, it, expect, vi } from 'vitest';
-import { DOMParser } from '@xmldom/xmldom';
-import JSZip from 'jszip';
-import { DocumentObject, Part, DocxPackage } from '../docx/bridge.js';
-import * as transforms from './transforms.js';
-import { finalize_document } from './core.js';
+import { describe, it, expect, vi } from "vitest";
+import { DOMParser } from "@xmldom/xmldom";
+import { DocumentObject, Part, DocxPackage } from "../docx/bridge.js";
+import * as transforms from "./transforms.js";
+import { finalize_document } from "./core.js";
 
 // --- Helper to build a lightweight in-memory DocumentObject ---
 function createMockDoc(bodyXml: string): DocumentObject {
   const fullXml = `<w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml"><w:body>${bodyXml}</w:body></w:document>`;
-  const doc = new DOMParser().parseFromString(fullXml, 'text/xml');
-  const zip = new JSZip();
-  const pkg = new DocxPackage(zip);
-  
-  const part = new Part('/word/document.xml', fullXml, doc.documentElement, 'application/xml');
+  const doc = new DOMParser().parseFromString(fullXml, "text/xml");
+  const unzipped: Record<string, Uint8Array> = {};
+  const pkg = new DocxPackage(unzipped);
+
+  const part = new Part(
+    "/word/document.xml",
+    fullXml,
+    doc.documentElement,
+    "application/xml",
+  );
   pkg.parts.push(part);
   pkg.mainDocumentPart = part;
-  
+
   return new DocumentObject(pkg, part);
 }
 
 // --- Transforms Unit Tests ---
-describe('Sanitize Transforms', () => {
-  
-  it('should strip RSID attributes and elements', () => {
+describe("Sanitize Transforms", () => {
+  it("should strip RSID attributes and elements", () => {
     const doc = createMockDoc(`
       <w:p w:rsidR="00A21F3B" w:rsidP="00B33E21">
         <w:r><w:t>Hello</w:t></w:r>
       </w:p>
       <w:sectPr><w:rsids><w:rsidRoot w:val="00A21F3B"/></w:rsids></w:sectPr>
     `);
-    
+
     const lines = transforms.strip_rsid(doc);
     const xml = doc.element.toString();
-    
+
     expect(lines.length).toBeGreaterThan(0);
-    expect(xml).not.toContain('w:rsidR');
-    expect(xml).not.toContain('w:rsidP');
-    expect(xml).not.toContain('w:rsids');
+    expect(xml).not.toContain("w:rsidR");
+    expect(xml).not.toContain("w:rsidP");
+    expect(xml).not.toContain("w:rsids");
   });
 
-  it('should strip w14:paraId and w14:textId', () => {
+  it("should strip w14:paraId and w14:textId", () => {
     const doc = createMockDoc(`
       <w:p w14:paraId="3F2A91BC" w14:textId="77777777">
         <w:r><w:t>Test</w:t></w:r>
       </w:p>
     `);
-    
+
     const lines = transforms.strip_para_ids(doc);
     const xml = doc.element.toString();
-    
+
     expect(lines.length).toBeGreaterThan(0);
-    expect(xml).not.toContain('w14:paraId');
-    expect(xml).not.toContain('w14:textId');
+    expect(xml).not.toContain("w14:paraId");
+    expect(xml).not.toContain("w14:textId");
   });
 
-  it('should strip hidden text runs', () => {
+  it("should strip hidden text runs", () => {
     const doc = createMockDoc(`
       <w:p>
         <w:r>
@@ -66,34 +69,43 @@ describe('Sanitize Transforms', () => {
         </w:r>
       </w:p>
     `);
-    
+
     const lines = transforms.strip_hidden_text(doc);
     const xml = doc.element.toString();
-    
+
     expect(lines.length).toBeGreaterThan(0);
-    expect(xml).not.toContain('HiddenSecret');
-    expect(xml).toContain('VisibleText');
+    expect(xml).not.toContain("HiddenSecret");
+    expect(xml).toContain("VisibleText");
   });
 
-  it('should scrub document properties', () => {
-    const doc = createMockDoc('<w:p/>');
-    
+  it("should scrub document properties", () => {
+    const doc = createMockDoc("<w:p/>");
+
     // Mock docProps/app.xml
-    const appXml = '<Properties><TotalTime>15</TotalTime><Template>Confidential.dotm</Template></Properties>';
-    const appEl = new DOMParser().parseFromString(appXml, 'text/xml').documentElement;
-    const appPart = new Part('/docProps/app.xml', appXml, appEl, 'application/xml');
+    const appXml =
+      "<Properties><TotalTime>15</TotalTime><Template>Confidential.dotm</Template></Properties>";
+    const appEl = new DOMParser().parseFromString(
+      appXml,
+      "text/xml",
+    ).documentElement;
+    const appPart = new Part(
+      "/docProps/app.xml",
+      appXml,
+      appEl,
+      "application/xml",
+    );
     doc.pkg.parts.push(appPart);
-    
+
     const lines = transforms.scrub_doc_properties(doc);
     const resultXml = appPart._element.toString();
-    
+
     expect(lines.length).toBeGreaterThan(0);
-    expect(resultXml).toContain('<TotalTime>0</TotalTime>');
-    expect(resultXml).toContain('<Template/>');
-    expect(resultXml).not.toContain('Confidential.dotm');
+    expect(resultXml).toContain("<TotalTime>0</TotalTime>");
+    expect(resultXml).toContain("<Template/>");
+    expect(resultXml).not.toContain("Confidential.dotm");
   });
 
-  it('should strip custom XML parts and data bindings', () => {
+  it("should strip custom XML parts and data bindings", () => {
     const doc = createMockDoc(`
       <w:p>
         <w:sdt>
@@ -101,19 +113,26 @@ describe('Sanitize Transforms', () => {
         </w:sdt>
       </w:p>
     `);
-    
+
     // Mock custom XML part
-    const customPart = new Part('/customXml/item1.xml', '<t/>', new DOMParser().parseFromString('<t/>', 'text/xml').documentElement, 'application/xml');
+    const customPart = new Part(
+      "/customXml/item1.xml",
+      "<t/>",
+      new DOMParser().parseFromString("<t/>", "text/xml").documentElement,
+      "application/xml",
+    );
     doc.pkg.parts.push(customPart);
-    
+
     const lines = transforms.strip_custom_xml(doc);
-    
+
     expect(lines.length).toBeGreaterThan(0);
-    expect(doc.pkg.parts.find(p => p.partname.includes('customXml'))).toBeUndefined();
-    expect(doc.element.toString()).not.toContain('w:dataBinding');
+    expect(
+      doc.pkg.parts.find((p) => p.partname.includes("customXml")),
+    ).toBeUndefined();
+    expect(doc.element.toString()).not.toContain("w:dataBinding");
   });
 
-  it('should count and accept all tracked changes', () => {
+  it("should count and accept all tracked changes", () => {
     const doc = createMockDoc(`
       <w:p>
         <w:del w:id="1">
@@ -124,74 +143,80 @@ describe('Sanitize Transforms', () => {
         </w:ins>
       </w:p>
     `);
-    
+
     const [ins, del, fmt] = transforms.count_tracked_changes(doc);
     expect(ins).toBe(1);
     expect(del).toBe(1);
-    
+
     const lines = transforms.accept_all_tracked_changes(doc);
     const xml = doc.element.toString();
-    
+
     expect(lines.length).toBeGreaterThan(0);
-    expect(xml).not.toContain('w:del');
-    expect(xml).not.toContain('w:ins');
-    expect(xml).not.toContain('Vendor'); // Deletion was removed
-    expect(xml).toContain('Supplier'); // Insertion was unwrapped
+    expect(xml).not.toContain("w:del");
+    expect(xml).not.toContain("w:ins");
+    expect(xml).not.toContain("Vendor"); // Deletion was removed
+    expect(xml).toContain("Supplier"); // Insertion was unwrapped
   });
-
 });
 
 // --- Orchestrator Integration Tests ---
-describe('Finalize Document (Core)', () => {
+describe("Finalize Document (Core)", () => {
+  it("should inject XML locking (Read-Only) into settings.xml", async () => {
+    const doc = createMockDoc("<w:p/>");
 
-  it('should inject XML locking (Read-Only) into settings.xml', async () => {
-    const doc = createMockDoc('<w:p/>');
-    
     // Mock word/settings.xml
-    const settingsXml = '<w:settings xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"></w:settings>';
-    const settingsEl = new DOMParser().parseFromString(settingsXml, 'text/xml').documentElement;
-    const settingsPart = new Part('/word/settings.xml', settingsXml, settingsEl, 'application/xml');
+    const settingsXml =
+      '<w:settings xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"></w:settings>';
+    const settingsEl = new DOMParser().parseFromString(
+      settingsXml,
+      "text/xml",
+    ).documentElement;
+    const settingsPart = new Part(
+      "/word/settings.xml",
+      settingsXml,
+      settingsEl,
+      "application/xml",
+    );
     doc.pkg.parts.push(settingsPart);
-    
+
     // Mock the doc.save buffer return
-    doc.save = vi.fn().mockResolvedValue(Buffer.from('mock'));
-    
-    const res = await finalize_document(doc, { 
-      filename: 'test.docx', 
-      protection_mode: 'read_only' 
+    doc.save = vi.fn().mockResolvedValue(Buffer.from("mock"));
+
+    const res = await finalize_document(doc, {
+      filename: "test.docx",
+      protection_mode: "read_only",
     });
-    
+
     const finalSettings = settingsPart._element.toString();
 
-    expect(res.reportText).toContain('Result: CLEAN');
-    expect(res.reportText).toContain('Document locked (Read-Only');
-    
+    expect(res.reportText).toContain("Result: CLEAN");
+    expect(res.reportText).toContain("Document locked (Read-Only");
+
     // Validate mathematical injection
-    expect(finalSettings).toContain('w:documentProtection');
+    expect(finalSettings).toContain("w:documentProtection");
     expect(finalSettings).toContain('w:edit="readOnly"');
     expect(finalSettings).toContain('w:enforcement="1"');
   });
 
-  it('should return a blocked status if unaccepted changes remain and accept_all is false', async () => {
+  it("should return a blocked status if unaccepted changes remain and accept_all is false", async () => {
     const doc = createMockDoc(`
       <w:p>
         <w:ins w:id="1"><w:r><w:t>Unresolved Edit</w:t></w:r></w:ins>
       </w:p>
     `);
-    
-    const res = await finalize_document(doc, { 
-      filename: 'draft.docx', 
-      sanitize_mode: 'full',
-      accept_all: false // <-- Should block
+
+    const res = await finalize_document(doc, {
+      filename: "draft.docx",
+      sanitize_mode: "full",
+      accept_all: false, // <-- Should block
     });
-    
-    expect(res.reportText).toContain('BLOCKED:');
-    expect(res.reportText).toContain('unresolved tracked changes');
+
+    expect(res.reportText).toContain("BLOCKED:");
+    expect(res.reportText).toContain("unresolved tracked changes");
   });
 
-  describe('Resolved Bugs Sanitize Parity Verification', () => {
-    
-    it('BUG-FRAG-1: Coalesces adjacent identical runs after accepting tracked changes', async () => {
+  describe("Resolved Bugs Sanitize Parity Verification", () => {
+    it("BUG-FRAG-1: Coalesces adjacent identical runs after accepting tracked changes", async () => {
       const doc = createMockDoc(`
         <w:p>
           <w:r><w:t xml:space="preserve">The term shall be </w:t></w:r>
@@ -199,39 +224,44 @@ describe('Finalize Document (Core)', () => {
           <w:r><w:t xml:space="preserve"> years from the Effective Date.</w:t></w:r>
         </w:p>
       `);
-      
-      doc.save = vi.fn().mockResolvedValue(Buffer.from('mock'));
-      
+
+      doc.save = vi.fn().mockResolvedValue(Buffer.from("mock"));
+
       await finalize_document(doc, {
-        filename: 'test.docx',
-        sanitize_mode: 'full',
-        accept_all: true
+        filename: "test.docx",
+        sanitize_mode: "full",
+        accept_all: true,
       });
 
       const xml = doc.element.toString();
       // We should see a single coalesced string rather than fragmented <w:t> nodes
-      expect(xml).toContain('The term shall be five (5) years from the Effective Date.');
+      expect(xml).toContain(
+        "The term shall be five (5) years from the Effective Date.",
+      );
 
-      const runs = doc.element.getElementsByTagName('w:r');
+      const runs = doc.element.getElementsByTagName("w:r");
       // If they are coalesced properly, there will be exactly 1 run instead of 3
       expect(runs.length).toBe(1);
     });
 
-    it('BUG-NS-1: Strips unused xmlns:w16du namespace declarations during finalization', async () => {
-      const doc = createMockDoc('<w:p/>');
+    it("BUG-NS-1: Strips unused xmlns:w16du namespace declarations during finalization", async () => {
+      const doc = createMockDoc("<w:p/>");
       // Manually inject the namespace onto the absolute root as the engine does
-      doc.part._element.setAttribute('xmlns:w16du', 'http://schemas.microsoft.com/office/word/2023/wordml/word16du');
-      
-      doc.save = vi.fn().mockResolvedValue(Buffer.from('mock'));
+      doc.part._element.setAttribute(
+        "xmlns:w16du",
+        "http://schemas.microsoft.com/office/word/2023/wordml/word16du",
+      );
+
+      doc.save = vi.fn().mockResolvedValue(Buffer.from("mock"));
 
       await finalize_document(doc, {
-        filename: 'test.docx',
-        sanitize_mode: 'full'
+        filename: "test.docx",
+        sanitize_mode: "full",
       });
 
       // The final stringified XML of the root document should NOT contain the unused namespace
       const xml = doc.part._element.toString();
-      expect(xml).not.toContain('xmlns:w16du');
+      expect(xml).not.toContain("xmlns:w16du");
     });
   });
-});
+});