Fix case sensitivity in Repo class URL handling (#107)

Previously, the class docbuild.models.repo.Repo model treated URLs
with different casing (e.g., ``github.com`` vs ``GitHub.com``) as unequal.

This commit normalizes URLs to ensure that casing differences do not
result in duplicate or distinct records for the same repository.

Author: tomschr

changelog.d/107.bugfix.rst

@@ -0,0 +1,6 @@
+Fix case sensitivity in :class:`~docbuild.models.repo.Repo` handling.
+
+Previously, the model treated URLs with different casing (e.g., ``github.com``
+vs ``GitHub.com``) as unequal. This commit normalizes URLs to ensure that
+casing differences do not result in duplicate or distinct records for the
+same repository.

src/docbuild/models/repo.py

@@ -78,9 +78,8 @@ def __init__(self, value: str) -> None:
         ssh_match = self._SSH_PATTERN.match(value)
 
         if 'https://' in value or 'http://' in value:
-            parsed_original = urlparse(value)
-            name = parsed_original.path.strip('/').lower().rsplit('.git', 1)[0]
-            name = name.rstrip('/')
+            parsed_original = urlparse(value.lower())
+            name = parsed_original.path.strip('/').rsplit('.git', 1)[0]
             url = f'{parsed_original.scheme}://{parsed_original.netloc}/{name}.git'
             host = f'{parsed_original.scheme}://{parsed_original.netloc}'
             surl = f'{self._MAP_URL2SERVICE.get(host, "gh")}://{name}'

tests/models/test_repo.py

@@ -3,9 +3,12 @@
 from docbuild.models.repo import Repo
 
 
-@pytest.fixture
-def repo_url() -> str:
-    return 'https://github.com/org/repo.git'
+# Use different spellings/casings of the same URL to test normalization
+@pytest.fixture(
+    params=['https://github.com/org/repo.git', 'https://GitHub.com/ORG/repo.git']
+)
+def repo_url(request) -> str:
+    return request.param
 
 
 @pytest.mark.parametrize(
@@ -29,6 +32,12 @@ def repo_url() -> str:
             'org/repo_git',
             'https://github.com/org/repo_git.git',
         ),
+        # 6
+        (
+            'https://GitHub.com/ORG/repo_git.git/',
+            'org/repo_git',
+            'https://github.com/org/repo_git.git',
+        ),
     ],
 )
 def test_repo_https(input_value, name, url):
@@ -54,6 +63,12 @@ def test_repo_https(input_value, name, url):
             'org/repo_git',
             'https://github.com/org/repo_git.git',
         ),
+        # 6
+        (
+            '[email protected]:org/repo_git/',
+            'org/repo_git',
+            'https://github.com/org/repo_git.git',
+        ),
     ],
 )
 def test_repo_ssh(input_value, name, url):
@@ -106,6 +121,8 @@ def test_repo_abbreviation(input_value, name, url):
         ('gh://org/repo.git', 'gh://org/repo'),
         # 8
         ('gh://ORG/repo_git.git', 'gh://org/repo_git'),
+        # 9
+        ('GH://ORG/repo_git.git', 'gh://org/repo_git'),
     ],
 )
 def test_repo_with_surl(input_value, surl):
@@ -177,4 +194,4 @@ def test_repo_surl(repo_url):
 
 def test_repo_str(repo_url):
     repo = Repo(repo_url)
-    assert str(repo) == repo_url
+    assert str(repo) == repo_url.lower()