fix: sanitize invalid UTF-8 bytes in uploaded filenames (#265)

* fix: sanitize invalid UTF-8 bytes in uploaded filenames

A client-supplied filename containing a raw non-UTF-8 byte (e.g. 0x97,
a Windows-1252 em dash) crashed the media insert with an uncaught
QueryException: Postgres rejects invalid UTF-8 byte sequences outright
under UTF8 encoding. Centralized a sanitizeOriginalFilename() helper
in HasMedia and applied it to all three insert paths that store
original_filename: addMedia(), addMediaFromPath(), and
addMediaFromStoredPath() (the multipart cloud-upload registration
path) — so every upload entry point is covered, not just the one that
happened to crash in production.

Fixes Nightwatch issue #24.

* refactor: use mb_scrub() instead of the mb_convert_encoding same-encoding trick

mb_scrub() (PHP 8.1+) is the purpose-built function for scrubbing
invalid byte sequences — same behavior, clearer intent than the
convert-to-same-encoding workaround it replaces.

* test: cover addMediaFromStoredPath (previously untested, including sanitize fix)

addMediaFromStoredPath — the multipart cloud-upload registration path
— had zero test coverage before this PR, including for the invalid
UTF-8 filename fix applied to it. Added a basic happy-path test plus
the sanitize regression test, matching the coverage already added for
addMedia() and addMediaFromPath().

Verified the regression test actually catches the bug: reverted the
sanitize call for this one method locally, confirmed the test fails
with the exact Nightwatch #24 QueryException, then restored the fix.

Author: paulocastellano

app/Models/Traits/HasMedia.php

@@ -98,7 +98,7 @@ public function addMedia(UploadedFile $file, string $collection = 'default', arr
             'collection' => $collection,
             'type' => $type,
             'path' => $path,
-            'original_filename' => $file->getClientOriginalName(),
+            'original_filename' => $this->sanitizeOriginalFilename($file->getClientOriginalName()),
             'mime_type' => $normalizedMime,
             'size' => strlen($normalizedBytes),
             'order' => 0,
@@ -137,7 +137,7 @@ public function addMediaFromPath(string $filePath, string $originalFilename, str
             'collection' => $collection,
             'type' => $type,
             'path' => $stored['path'],
-            'original_filename' => $originalFilename,
+            'original_filename' => $this->sanitizeOriginalFilename($originalFilename),
             'mime_type' => $stored['mime_type'],
             'size' => $stored['size'],
             'order' => 0,
@@ -169,7 +169,7 @@ public function addMediaFromStoredPath(
             'collection' => $collection,
             'type' => $type,
             'path' => $storagePath,
-            'original_filename' => $originalFilename,
+            'original_filename' => $this->sanitizeOriginalFilename($originalFilename),
             'mime_type' => $mimeType,
             'size' => $size,
             'order' => 0,
@@ -251,6 +251,16 @@ private function getMediaType(string $mimeType): string
             ?? throw new InvalidArgumentException("Unsupported media MIME type: {$mimeType}"))->value;
     }
 
+    /**
+     * Client-supplied filenames may contain byte sequences that aren't valid
+     * UTF-8 (e.g. a raw Windows-1252 byte for an em dash). Postgres rejects
+     * those outright on insert, so replace invalid sequences before storing.
+     */
+    private function sanitizeOriginalFilename(string $filename): string
+    {
+        return mb_scrub($filename, 'UTF-8');
+    }
+
     private function getMediaMeta(UploadedFile $file, string $type): array
     {
         $meta = [];

tests/Feature/Api/UploadControllerTest.php

@@ -53,6 +53,24 @@ function signedUploadUrl(Workspace $ws, string $token, ?int $expiresInMinutes =
         ]);
 });
 
+test('sanitizes an invalid UTF-8 byte in the client filename instead of crashing the insert (Nightwatch #24)', function () {
+    $token = (string) Str::uuid();
+    // 0x97 is a raw Windows-1252 em dash, not valid UTF-8 on its own — Postgres
+    // rejects it outright on insert unless the filename is sanitized first.
+    $file = UploadedFile::fake()->image("earnings \x97 report.png", 50, 50);
+
+    $response = $this->post(signedUploadUrl($this->workspace, $token), [
+        'media' => $file,
+    ]);
+
+    $response->assertCreated();
+
+    $media = Media::where('upload_token', $token)->first();
+    expect($media)->not->toBeNull();
+    expect(mb_check_encoding($media->original_filename, 'UTF-8'))->toBeTrue();
+    expect($media->original_filename)->toBe('earnings ? report.png');
+});
+
 test('rejects unsigned request', function () {
     $token = (string) Str::uuid();
     $file = UploadedFile::fake()->image('shot.png', 50, 50);

tests/Unit/Traits/HasMediaTest.php

@@ -324,6 +324,57 @@
         ->and(pathinfo($media->path, PATHINFO_EXTENSION))->toBe('jpg');
 });
 
+test('model can add media from stored path', function () {
+    $workspace = Workspace::factory()->create();
+    $content = file_get_contents(__DIR__.'/../../fixtures/1x1.png');
+    Storage::put('medias/existing.png', $content);
+
+    $media = $workspace->addMediaFromStoredPath('medias/existing.png', 'existing.png', 'image/png', strlen($content), 'assets');
+
+    expect($media)->toBeInstanceOf(Media::class);
+    expect($media->original_filename)->toBe('existing.png');
+    expect($media->path)->toBe('medias/existing.png');
+    expect($media->mime_type)->toBe('image/png');
+    expect($media->size)->toBe(strlen($content));
+});
+
+test('add media from stored path sanitizes invalid UTF-8 bytes in the original filename', function () {
+    $workspace = Workspace::factory()->create();
+    $content = file_get_contents(__DIR__.'/../../fixtures/1x1.png');
+    Storage::put('medias/existing.png', $content);
+    $invalidName = "earnings \x97 report.png";
+
+    $media = $workspace->addMediaFromStoredPath('medias/existing.png', $invalidName, 'image/png', strlen($content), 'assets');
+
+    expect(mb_check_encoding($media->original_filename, 'UTF-8'))->toBeTrue();
+    expect($media->original_filename)->toBe('earnings ? report.png');
+});
+
+test('add media sanitizes invalid UTF-8 bytes in the original filename', function () {
+    $workspace = Workspace::factory()->create();
+    $invalidName = "earnings \x97 report.jpg";
+    $file = UploadedFile::fake()->image($invalidName, 100, 100);
+
+    $media = $workspace->addMedia($file, 'assets');
+
+    expect(mb_check_encoding($media->original_filename, 'UTF-8'))->toBeTrue();
+    expect($media->original_filename)->toBe('earnings ? report.jpg');
+});
+
+test('add media from path sanitizes invalid UTF-8 bytes in the original filename', function () {
+    $workspace = Workspace::factory()->create();
+    $tempFile = tempnam(sys_get_temp_dir(), 'test');
+    file_put_contents($tempFile, file_get_contents(__DIR__.'/../../fixtures/1x1.png'));
+    $invalidName = "earnings \x97 report.png";
+
+    $media = $workspace->addMediaFromPath($tempFile, $invalidName, 'assets');
+
+    expect(mb_check_encoding($media->original_filename, 'UTF-8'))->toBeTrue();
+    expect($media->original_filename)->toBe('earnings ? report.png');
+
+    unlink($tempFile);
+});
+
 test('client meta is merged into media meta', function () {
     $workspace = Workspace::factory()->create();
     $file = UploadedFile::fake()->image('photo.jpg', 640, 480);