fix: add binary regression tests to Base64Encoder and Blob.load() using a fixture with non-UTF-8 byte sequences to guard against silent data corruption

Author: ryancyq

.gitignore

@@ -1,5 +1,4 @@
 .envrc
-.idea/
 
 # Logs
 logs

__tests__/blob.test.ts

@@ -72,6 +72,14 @@ describe('Blob', () => {
       )
     })
 
+    it('binary file without corruption', async () => {
+      const blob = getBlob('fixtures/blob.bin')
+      const fileAddition = await blob.load()
+      expect(fileAddition.contents).toEqual(
+        fs.readFileSync(join(__dirname, 'fixtures/blob.bin.base64.txt')).toString()
+      )
+    })
+
     it('file with string', async () => {
       const blob = getBlob('fixtures/error.txt')
       const mockStream = new PassThrough()

__tests__/fixtures/blob.bin

@@ -0,0 +1 @@
+iVBORw0KGgr//gAB

__tests__/fixtures/blob.bin.base64.txt

@@ -15,4 +15,16 @@ describe('Base64 Encoder', () => {
     const streamedContent = Buffer.concat(chunks).toString('utf8')
     expect(streamedContent).toEqual(Buffer.from(content).toString('base64'))
   })
+
+  it('binary stream without corruption', async () => {
+    const binary = Buffer.from([0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0xff, 0xfe, 0x00, 0x01])
+    const stream = Readable.from(binary).pipe(new Base64Encoder())
+
+    const chunks: Buffer[] = []
+    for await (const chunk of stream) {
+      chunks.push(Buffer.isBuffer(chunk) ? chunk : Buffer.from(chunk))
+    }
+    const streamedContent = Buffer.concat(chunks).toString('utf8')
+    expect(streamedContent).toEqual(binary.toString('base64'))
+  })
 })

__tests__/stream/base64-encoder.test.ts

@@ -26,9 +26,6 @@ export class Blob {
       throw new Error(`File does not exist, path: ${this.absolutePath}`)
     }
 
-    // Always read files as raw buffers without encoding
-    // The Base64Encoder works with buffers for both text and binary files
-    // Using any encoding (like 'utf8') corrupts the data
     return fs
       .createReadStream(this.absolutePath)
       .pipe(new Base64Encoder())