fix(worker): download_asset支持HTTP URL下载,兼容预签名file_url #487

Merged
auto-approve-bot merged 1 commits from fix/download-asset-http-fallback into develop 2026-07-17 20:46:29 +08:00
2 changed files with 218 additions and 1 deletions
+29 -1
View File
@@ -14,6 +14,7 @@ from pathlib import Path
from urllib.parse import urlparse
import oss2
import requests
from packages.shared.config import get_shared_settings
@@ -96,13 +97,22 @@ def normalize_storage_key(storage_key_or_url: str) -> str:
def download_asset(asset_storage_key: str, local_path: Path) -> bool:
"""从 OSS 下载素材文件到本地路径。
自动识别输入类型:
- 完整 URLhttp:// 或 https:// 开头)→ 走 HTTP 下载(支持预签名URL)
- OSS 存储键 → 走 oss2 SDK 下载
Args:
asset_storage_key: 素材的存储键或完整 URL
asset_storage_key: 素材的存储键或完整 URL
local_path: 本地保存路径
Returns:
True 表示下载成功,False 表示失败。
"""
# 完整URL走HTTP下载(兼容预签名URL)
if asset_storage_key.startswith(("http://", "https://")):
return _download_via_http(asset_storage_key, local_path)
# OSS存储键走SDK
bucket = oss_bucket()
if bucket is None:
return False
@@ -114,6 +124,24 @@ def download_asset(asset_storage_key: str, local_path: Path) -> bool:
return False
def _download_via_http(url: str, local_path: Path) -> bool:
"""通过 HTTP 下载文件(支持预签名 URL)。
使用流式下载避免大文件内存溢出,超时 300s。
"""
try:
resp = requests.get(url, stream=True, timeout=300)
resp.raise_for_status()
with open(local_path, "wb") as f:
for chunk in resp.iter_content(chunk_size=8 * 1024 * 1024):
if chunk:
f.write(chunk)
return local_path.exists() and local_path.stat().st_size > 0
except Exception:
logger.exception("HTTP下载素材失败: %s", url)
return False
def upload_to_oss(local_path: Path, storage_key: str) -> str | None:
"""上传文件到 OSS,返回公开 URL。
+189
View File
@@ -0,0 +1,189 @@
"""oss_helpers.download_asset 单元测试 — HTTP URL / OSS key 双路径覆盖."""
from __future__ import annotations
from pathlib import Path
from unittest.mock import MagicMock, patch
import pytest
@pytest.fixture
def tmp_asset_dir(tmp_path: Path) -> Path:
d = tmp_path / "assets"
d.mkdir()
return d
# ── HTTP URL 下载路径 ──────────────────────────────────────────────────────
class TestDownloadAssetHttpUrl:
"""传入完整 https:// URL 时走 requests HTTP 下载,不走 oss2 SDK."""
@patch("apps.worker.video_processing.oss_helpers.requests.get")
def test_http_url_download_success(self, mock_get, tmp_asset_dir):
"""HTTP URL 下载成功."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_resp = MagicMock()
mock_resp.status_code = 200
mock_resp.iter_content.return_value = [b"fake-video-data" * 100]
mock_get.return_value = mock_resp
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("https://oss.example.com/assets/test.mp4?sign=xxx", local_path)
assert result is True
assert local_path.exists()
assert local_path.stat().st_size > 0
mock_get.assert_called_once()
# 不走 oss2 bucket
assert "oss_bucket" not in str(mock_get.call_args)
@patch("apps.worker.video_processing.oss_helpers.requests.get")
def test_http_url_download_403_returns_false(self, mock_get, tmp_asset_dir):
"""HTTP URL 返回 403,下载失败返回 False."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_resp = MagicMock()
mock_resp.status_code = 403
mock_resp.raise_for_status.side_effect = Exception("403 Forbidden")
mock_get.return_value = mock_resp
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("https://oss.example.com/assets/expired.mp4?sign=old", local_path)
assert result is False
@patch("apps.worker.video_processing.oss_helpers.requests.get")
def test_http_url_connection_error_returns_false(self, mock_get, tmp_asset_dir):
"""HTTP URL 连接超时,返回 False."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_get.side_effect = Exception("Connection timeout")
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("https://unreachable.example.com/test.mp4", local_path)
assert result is False
@patch("apps.worker.video_processing.oss_helpers.requests.get")
def test_http_url_streams_in_chunks(self, mock_get, tmp_asset_dir):
"""HTTP 下载使用流式 + 8MB chunk,避免大文件内存溢出."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_resp = MagicMock()
mock_resp.status_code = 200
mock_resp.iter_content.return_value = [b"chunk1", b"chunk2", b"chunk3"]
mock_get.return_value = mock_resp
local_path = tmp_asset_dir / "test.mp4"
download_asset("https://oss.example.com/assets/big.mp4", local_path)
mock_resp.iter_content.assert_called_once()
call_kwargs = mock_resp.iter_content.call_args
assert call_kwargs.kwargs.get("chunk_size") == 8 * 1024 * 1024
# ── OSS key 下载路径 ──────────────────────────────────────────────────────
class TestDownloadAssetOssKey:
"""传入 OSS 存储键时走 oss2 SDK,不走 HTTP."""
@patch("apps.worker.video_processing.oss_helpers.requests.get")
@patch("apps.worker.video_processing.oss_helpers.oss_bucket")
def test_oss_key_uses_sdk_not_http(self, mock_oss_bucket, mock_requests_get, tmp_asset_dir):
"""OSS key 路径走 oss2 SDK,不调用 requests.get."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_bucket = MagicMock()
mock_bucket.get_object_to_file.side_effect = lambda key, path: Path(path).write_bytes(b"oss-data")
mock_oss_bucket.return_value = mock_bucket
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("assets/user123/test.mp4", local_path)
assert result is True
mock_bucket.get_object_to_file.assert_called_once()
mock_requests_get.assert_not_called()
@patch("apps.worker.video_processing.oss_helpers.oss_bucket")
def test_oss_bucket_none_returns_false(self, mock_oss_bucket, tmp_asset_dir):
"""OSS 未配置时返回 False."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_oss_bucket.return_value = None
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("assets/test.mp4", local_path)
assert result is False
@patch("apps.worker.video_processing.oss_helpers.oss_bucket")
def test_oss_sdk_exception_returns_false(self, mock_oss_bucket, tmp_asset_dir):
"""OSS SDK 抛异常时返回 False,不崩溃."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_bucket = MagicMock()
mock_bucket.get_object_to_file.side_effect = Exception("NoSuchKey")
mock_oss_bucket.return_value = mock_bucket
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("assets/nonexistent.mp4", local_path)
assert result is False
# ── 路径选择逻辑 ──────────────────────────────────────────────────────────
class TestDownloadAssetRouteSelection:
"""URL vs OSS key 的路由选择."""
@patch("apps.worker.video_processing.oss_helpers._download_via_http")
@patch("apps.worker.video_processing.oss_helpers.oss_bucket")
def test_https_url_routes_to_http(self, mock_oss_bucket, mock_http_download, tmp_asset_dir):
"""https:// 开头走 HTTP 下载."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_http_download.return_value = True
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("https://oss.example.com/a/b.mp4", local_path)
assert result is True
mock_http_download.assert_called_once()
mock_oss_bucket.assert_not_called()
@patch("apps.worker.video_processing.oss_helpers._download_via_http")
@patch("apps.worker.video_processing.oss_helpers.oss_bucket")
def test_http_url_routes_to_http(self, mock_oss_bucket, mock_http_download, tmp_asset_dir):
"""http:// 开头也走 HTTP 下载."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_http_download.return_value = True
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("http://oss.example.com/a/b.mp4", local_path)
assert result is True
mock_http_download.assert_called_once()
mock_oss_bucket.assert_not_called()
@patch("apps.worker.video_processing.oss_helpers._download_via_http")
@patch("apps.worker.video_processing.oss_helpers.oss_bucket")
def test_plain_key_routes_to_oss_sdk(self, mock_oss_bucket, mock_http_download, tmp_asset_dir):
"""普通路径走 oss2 SDK."""
from apps.worker.video_processing.oss_helpers import download_asset
mock_bucket = MagicMock()
mock_bucket.get_object_to_file.side_effect = lambda key, path: Path(path).write_bytes(b"x")
mock_oss_bucket.return_value = mock_bucket
local_path = tmp_asset_dir / "test.mp4"
result = download_asset("user/123/asset.mp4", local_path)
assert result is True
mock_http_download.assert_not_called()