Improve the support for SourgeForge download URLs #26

CHANGELOG.rst

@@ -6,6 +6,9 @@ Release notes
 - Improve the stability of the "Check for new Package versions" feature.
   https://github.com/nexB/dejacode/issues/17
 
+- Improve the support for SourgeForge download URLs.
+  https://github.com/nexB/dejacode/issues/26
+
 ### Version 5.0.0
 
 Initial release.

component_catalog/tests/test_admin.py

@@ -2465,7 +2465,10 @@ def test_package_changeform_save_and_collect_data_on_addition(self, mock_get):
         }
 
         mock_get.return_value = mock.Mock(
-            content=b"\x00", headers={"content-length": 1}, status_code=200
+            content=b"\x00",
+            headers={"content-length": 1},
+            status_code=200,
+            url="http://domain.com/a.zip",
         )
 
         response = self.client.post(add_url, data, follow=True)

component_catalog/tests/test_models.py

@@ -2213,7 +2213,7 @@ def test_collect_package_data(self, mock_get):
         with self.assertRaisesMessage(DataCollectionException, expected_message):
             collect_package_data("ftp://ftp.denx.de/pub/u-boot/u-boot-2017.11.tar.bz2")
 
-        package_url = "http://domain.com/a.zip;<params>?<query>#<fragment>"
+        download_url = "http://domain.com/a%20b.zip;<params>?<query>#<fragment>"
 
         default_max_length = download.CONTENT_MAX_LENGTH
         download.CONTENT_MAX_LENGTH = 0
@@ -2223,15 +2223,18 @@ def test_collect_package_data(self, mock_get):
             content=b"\x00", headers={"content-length": 300000000}, status_code=200
         )
         with self.assertRaisesMessage(DataCollectionException, expected_message):
-            collect_package_data(package_url)
+            collect_package_data(download_url)
 
         download.CONTENT_MAX_LENGTH = default_max_length
         mock_get.return_value = mock.Mock(
-            content=b"\x00", headers={"content-length": 1}, status_code=200
+            content=b"\x00",
+            headers={"content-length": 1},
+            status_code=200,
+            url=download_url,
         )
         expected_data = {
-            "download_url": "http://domain.com/a.zip;<params>?<query>#<fragment>",
-            "filename": "a.zip",
+            "download_url": download_url,
+            "filename": "a b.zip",
             "size": 1,
             "sha1": "5ba93c9db0cff93f52b521d7420e43f6eda2784f",
             "md5": "93b885adfe0da089cdf634904fd59f71",
@@ -2241,7 +2244,7 @@ def test_collect_package_data(self, mock_get):
                 "4a802a71c3580b6370de4ceb293c324a8423342557d4e5c38438f0e36910ee"
             ),
         }
-        self.assertEqual(expected_data, collect_package_data(package_url))
+        self.assertEqual(expected_data, collect_package_data(download_url))
 
         expected_message = (
             "Exception Value: HTTPConnectionPool"
@@ -2253,15 +2256,15 @@ def test_collect_package_data(self, mock_get):
         )
         mock_get.return_value = response
         with self.assertRaisesMessage(DataCollectionException, expected_message):
-            collect_package_data(package_url)
+            collect_package_data(download_url)
 
         headers = {
             "content-length": 1,
             "content-disposition": 'attachment; filename="another_name.zip"',
         }
         mock_get.return_value = mock.Mock(content=b"\x00", headers=headers, status_code=200)
         expected_data = {
-            "download_url": "http://domain.com/a.zip;<params>?<query>#<fragment>",
+            "download_url": download_url,
             "filename": "another_name.zip",
             "size": 1,
             "sha1": "5ba93c9db0cff93f52b521d7420e43f6eda2784f",
@@ -2272,7 +2275,7 @@ def test_collect_package_data(self, mock_get):
                 "4a802a71c3580b6370de4ceb293c324a8423342557d4e5c38438f0e36910ee"
             ),
         }
-        self.assertEqual(expected_data, collect_package_data(package_url))
+        self.assertEqual(expected_data, collect_package_data(download_url))
 
     def test_package_create_save_set_usage_policy_from_license(self):
         from policy.models import AssociatedPolicy

dejacode_toolkit/download.py

@@ -7,8 +7,9 @@
 #
 
 import cgi
-import os
 import socket
+from pathlib import Path
+from urllib.parse import unquote
 from urllib.parse import urlparse
 
 from django.template.defaultfilters import filesizeformat
@@ -29,7 +30,7 @@ class DataCollectionException(Exception):
 
 def collect_package_data(url):
     try:
-        response = requests.get(url, timeout=10, stream=True)
+        response = requests.get(url, timeout=5, stream=True)
     except (requests.RequestException, socket.timeout) as e:
         raise DataCollectionException(e)
 
@@ -54,8 +55,13 @@ def collect_package_data(url):
         )
 
     content_disposition = response.headers.get("content-disposition", "")
-    value, params = cgi.parse_header(content_disposition)
-    filename = params.get("filename") or os.path.basename(urlparse(url).path)
+    _, params = cgi.parse_header(content_disposition)
+
+    filename = params.get("filename")
+    if not filename:
+        # Using ``response.url`` in place of provided ``url`` arg since the former
+        # will be more accurate in case of HTTP redirect.
+        filename = unquote(Path(urlparse(response.url).path).name)
 
     package_data = {
         "download_url": url,

setup.cfg

@@ -155,7 +155,7 @@ install_requires =
     openpyxl==3.1.2
     et-xmlfile==1.1.0
     # PackageURL
-    packageurl-python==0.12.0
+    packageurl-python==0.13.4
     # Gunicorn
     gunicorn==21.2.0
     # SPDX validation

thirdparty/dist/packageurl_python-0.13.4-py3-none-any.whl.ABOUT

@@ -0,0 +1,12 @@
+about_resource: packageurl_python-0.13.4-py3-none-any.whl
+name: packageurl-python
+version: 0.13.4
+download_url: https://files.pythonhosted.org/packages/6f/d6/dc41590e65a95198ad7490ed0fb34a1148e8eb5032c35c8d157b55aa496d/packageurl_python-0.13.4-py3-none-any.whl
+package_url: pkg:pypi/[email protected]
+license_expression: mit
+copyright: Copyright packageurl-python project contributors
+attribute: yes
+licenses:
+  - key: mit
+    name: MIT License
+    file: mit.LICENSE