feat(archive): add rar extraction support

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

package-lock.json

@@ -28,6 +28,7 @@
         "mammoth": "^1.12.0",
         "markdown-it": "^15.0.0",
         "multer": "^2.2.0",
+        "node-unrar-js": "^2.0.2",
         "opentype.js": "^2.0.0",
         "pdf-lib": "^1.17.1",
         "pdfjs-dist": "^6.2.108",
@@ -5307,6 +5308,15 @@
       "devOptional": true,
       "license": "MIT"
     },
+    "node_modules/node-unrar-js": {
+      "version": "2.0.2",
+      "resolved": "https://registry.npmjs.org/node-unrar-js/-/node-unrar-js-2.0.2.tgz",
+      "integrity": "sha512-hLNmoJzqaKJnod8yiTVGe9hnlNRHotUi0CreSv/8HtfRi/3JnRC8DvsmKfeGGguRjTEulhZK6zXX5PXoVuDZ2w==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=10.0.0"
+      }
+    },
     "node_modules/normalize-path": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/normalize-path/-/normalize-path-3.0.0.tgz",

package.json

@@ -35,6 +35,7 @@
     "mammoth": "^1.12.0",
     "markdown-it": "^15.0.0",
     "multer": "^2.2.0",
+    "node-unrar-js": "^2.0.2",
     "opentype.js": "^2.0.0",
     "pdf-lib": "^1.17.1",
     "pdfjs-dist": "^6.2.108",

src/converters/archive.js

@@ -6,6 +6,7 @@ import AdmZip from 'adm-zip';
 import { ZipArchive } from 'archiver';
 import * as tar from 'tar';
 import _7z from '7zip-min';
+import { createExtractorFromFile } from 'node-unrar-js';
 import { register } from './registry.js';
 
 const MAX_EXTRACTED_BYTES = 2 * 1024 * 1024 * 1024;
@@ -133,6 +134,15 @@ async function createTar7z(srcDir, outputPath, quality) {
   }
 }
 
+async function extractRar(inputPath, destDir) {
+  const extractor = await createExtractorFromFile({ filepath: inputPath, targetPath: destDir });
+  const { files } = extractor.extract();
+  for (const _file of files) {
+    // Iterating fully is required: node-unrar-js writes each entry to disk
+    // lazily as this generator is advanced.
+  }
+}
+
 const EXTRACTORS = {
   zip: extractZip,
   tar: extractTarLike,
@@ -140,6 +150,7 @@ const EXTRACTORS = {
   'tar.bz2': extractTarBz2,
   '7z': extract7z,
   'tar.7z': extractTar7z,
+  rar: extractRar,
 };
 const CREATORS = {
   zip: createZip,

test/converters/archive.test.js

@@ -228,3 +228,32 @@ describe('archive converters — 7z and tar.7z', () => {
     20000
   );
 });
+
+describe('archive converters — rar (extraction only)', () => {
+  it('lists rar as a source with the other 6 archive formats as targets, and never lists rar as a target for anything', () => {
+    const targets = listTargetFormats('rar');
+    expect(targets.sort()).toEqual(['7z', 'tar', 'tar.7z', 'tar.bz2', 'tar.gz', 'zip'].sort());
+    expect(listTargetFormats('zip')).not.toContain('rar');
+  });
+
+  it('converts the sample.rar fixture to zip', async () => {
+    const rarPath = path.join(import.meta.dirname, '..', 'fixtures', 'sample.rar');
+    const outputPath = path.join(tmpDir, 'from-rar.zip');
+    const entry = resolve('rar', 'zip');
+
+    await entry.convert(rarPath, outputPath);
+
+    const outZip = new AdmZip(outputPath);
+    expect(outZip.getEntries().length).toBeGreaterThan(0);
+  });
+});
+
+describe('archive converters — full registration matrix', () => {
+  it('registers all 42 source/target pairs (7 sources x 6 targets)', () => {
+    const sources = ['zip', 'tar', 'tar.gz', 'tar.bz2', '7z', 'tar.7z', 'rar'];
+    const targets = ['zip', 'tar', 'tar.gz', 'tar.bz2', '7z', 'tar.7z'];
+    for (const source of sources) {
+      expect(listTargetFormats(source).sort()).toEqual(targets.sort());
+    }
+  });
+});

test/fixtures/RAR_ATTRIBUTION.md

@@ -0,0 +1,9 @@
+# RAR fixture attribution
+
+`sample.rar` is a byte-identical copy of `testFiles/FolderTest.rar` from the
+[YuJianrong/node-unrar.js](https://github.com/YuJianrong/node-unrar.js)
+repository (MIT-licensed), used there for that project's own extraction
+tests. Reused here because there is no way to create a `.rar` fixture from
+this codebase — the `unrar` library's license bars any tool (including this
+project's own converters) from writing the RAR format, so RAR support is
+extraction-only everywhere, including in test fixtures.