More restrictive VCS URL validation (HTTPS only).

Signed-off-by: ziad hany <ziadhany2016@gmail.com>

Author: ziadhany

scanpipe/pipelines/scan_repo_grimoirelab.py

@@ -128,12 +128,8 @@ def is_valid_vcs_url(url):
         return False
 
     forbidden_chars = ["|", ";", "&", "`", "$(", ">", "<", "&&", "||"]
-    if any(char in forbidden_chars for char in url):
+    if any(char in url for char in forbidden_chars):
         return False
 
     parsed = urllib.parse.urlparse(url)
-    valid_schemes = {"https", "git", "ssh", "git+https", "git+ssh"}
-    if parsed.scheme in valid_schemes and parsed.netloc:
-        return True
-
-    return False
+    return parsed.scheme == "https" and bool(parsed.netloc)

scanpipe/tests/pipes/test_scan_repo_grimoirelab.py

@@ -104,11 +104,11 @@ def test_is_valid_vcs_url(self):
         test_cases = [
             # Valid URLs
             ("https://github.com/example/repo.git", True),
-            ("git://github.com/example/repo.git", True),
-            ("ssh://git@github.com/example/repo.git", True),
-            ("git+https://github.com/example/repo.git", True),
-            ("git+ssh://git@github.com/example/repo.git", True),
             # Invalid URLs
+            ("git://github.com/example/repo.git", False),
+            ("ssh://git@github.com/example/repo.git", False),
+            ("git+https://github.com/example/repo.git", False),
+            ("git+ssh://git@github.com/example/repo.git", False),
             ("git@github.com:example/repo.git", False),
             ("user@localhost:path/to/repo.git", False),
             ("hg@bitbucket.org:owner/repo", False),
@@ -118,6 +118,9 @@ def test_is_valid_vcs_url(self):
             ("git@github.com:repo.git\nrm -rf /", False),
             ("https://github.com/ repo.git", False),
             (" https://github.com/repo.git", False),
+            ("https://", False),
+            ("https:///repo.git", False),
+            ("https://?foo=bar", False),
             ("", False),
             (None, False),
         ]