refactor(crawler): 调整oss路径结构以统一目录层级

- 修改路径顺序，将平台(platform)调整至类别(category)之后

@@ -108,4 +108,4 @@ def build_oss_key(platform: str, category: str, filename: str) -> str:
     category: 'audio' | 'cover' | 'singer' | 'album'
     filename: 带扩展名的文件名
     """
-    return f"crawler/{platform}/{category}/{filename}"
+    return f"crawler/{category}/{platform}/{filename}"

@@ -113,7 +113,7 @@ def upload_plain_lyric_to_bucket(platform: str, unique_id: str, lyric: str, buck
     plain_lyric = extract_plain_lyric(lyric)
     if not plain_lyric:
         return ""
-    oss_key = f"crawler/{platform}/lyric/{unique_id}.txt"
+    oss_key = f"crawler/lyric/{platform}/{unique_id}.txt"
     bucket.put_object(oss_key, plain_lyric.encode("utf-8"), headers={'Content-Type': 'text/plain; charset=utf-8'})
     return f"{base_url.rstrip('/')}/{oss_key}"
 

@@ -42,9 +42,9 @@ def test_external_url_downloads_and_uploads():
     with patch('etl_to_crawler.oss._http_get') as mock_get:
         mock_get.return_value.content = fake_content
         mock_get.return_value.raise_for_status = MagicMock()
-        result = transfer_url(OTHER_URL, "crawler/qq/audio/abc.mp3", bucket, BASE_URL)
-    bucket.put_object.assert_called_once_with("crawler/qq/audio/abc.mp3", fake_content)
-    assert result == f"{BASE_URL}/crawler/qq/audio/abc.mp3"
+        result = transfer_url(OTHER_URL, "crawler/audio/qq/abc.mp3", bucket, BASE_URL)
+    bucket.put_object.assert_called_once_with("crawler/audio/qq/abc.mp3", fake_content)
+    assert result == f"{BASE_URL}/crawler/audio/qq/abc.mp3"
 
 
 def test_external_url_can_rewrite_download_base_to_internal_endpoint():
@@ -59,11 +59,11 @@ def test_external_url_can_rewrite_download_base_to_internal_endpoint():
         with patch('etl_to_crawler.oss._http_get') as mock_get:
             mock_get.return_value.content = fake_content
             mock_get.return_value.raise_for_status = MagicMock()
-            result = transfer_url(public_source, "crawler/qq/audio/abc.mp3", bucket, BASE_URL)
+            result = transfer_url(public_source, "crawler/audio/qq/abc.mp3", bucket, BASE_URL)
 
     mock_get.assert_called_once_with(f"{internal_source_base}/audio/abc.mp3", timeout=30)
-    bucket.put_object.assert_called_once_with("crawler/qq/audio/abc.mp3", fake_content)
-    assert result == f"{BASE_URL}/crawler/qq/audio/abc.mp3"
+    bucket.put_object.assert_called_once_with("crawler/audio/qq/abc.mp3", fake_content)
+    assert result == f"{BASE_URL}/crawler/audio/qq/abc.mp3"
 
 
 def test_transfer_url_with_md5_hashes_downloaded_content_before_upload():
@@ -72,9 +72,9 @@ def test_transfer_url_with_md5_hashes_downloaded_content_before_upload():
     with patch('etl_to_crawler.oss._http_get') as mock_get:
         mock_get.return_value.content = fake_content
         mock_get.return_value.raise_for_status = MagicMock()
-        result, audio_md5 = transfer_url_with_md5(OTHER_URL, "crawler/qq/audio/abc.mp3", bucket, BASE_URL)
-    bucket.put_object.assert_called_once_with("crawler/qq/audio/abc.mp3", fake_content)
-    assert result == f"{BASE_URL}/crawler/qq/audio/abc.mp3"
+        result, audio_md5 = transfer_url_with_md5(OTHER_URL, "crawler/audio/qq/abc.mp3", bucket, BASE_URL)
+    bucket.put_object.assert_called_once_with("crawler/audio/qq/abc.mp3", fake_content)
+    assert result == f"{BASE_URL}/crawler/audio/qq/abc.mp3"
     assert audio_md5 == "04d43544b267629d9089eaed3b847a99"
 
 
@@ -84,11 +84,11 @@ def test_audio_transfer_retries_timeout_then_uploads():
     response.raise_for_status.return_value = None
     with patch('etl_to_crawler.oss._http_get', side_effect=[requests.ReadTimeout('slow'), response]) as mock_get:
         with patch('etl_to_crawler.oss.time.sleep') as mock_sleep:
-            result, _ = transfer_url_with_md5(OTHER_URL, 'crawler/qq/audio/abc.mp3', bucket, BASE_URL)
+            result, _ = transfer_url_with_md5(OTHER_URL, 'crawler/audio/qq/abc.mp3', bucket, BASE_URL)
 
     assert mock_get.call_count == 2
     mock_sleep.assert_called_once()
-    assert result == f"{BASE_URL}/crawler/qq/audio/abc.mp3"
+    assert result == f"{BASE_URL}/crawler/audio/qq/abc.mp3"
 
 
 def test_transfer_does_not_retry_not_found_response():
@@ -98,7 +98,7 @@ def test_transfer_does_not_retry_not_found_response():
     with patch('etl_to_crawler.oss._http_get', side_effect=error) as mock_get:
         with patch('etl_to_crawler.oss.time.sleep') as mock_sleep:
             try:
-                transfer_url_with_md5(OTHER_URL, 'crawler/qq/audio/abc.mp3', bucket, BASE_URL)
+                transfer_url_with_md5(OTHER_URL, 'crawler/audio/qq/abc.mp3', bucket, BASE_URL)
             except requests.HTTPError:
                 pass
             else:

@@ -522,5 +522,5 @@ def test_process_netease_keeps_timestamped_lyric_and_uploads_plain_lyric(monkeyp
 
     assert inserted_songs[0]['lyric'] == '[ti:歌名]\n[00:01.00]第一句\n[00:02.00]第二句'
     assert inserted_songs[0]['audio_md5'] == 'audio-md5'
-    assert inserted_songs[0]['lyric_url'] == 'https://bucket.example.com/crawler/netease/lyric/300.txt'
+    assert inserted_songs[0]['lyric_url'] == 'https://bucket.example.com/crawler/lyric/netease/300.txt'
     assert uploaded['body'].decode('utf-8') == '第一句\n第二句'