fix: strip query/fragment from repo URLs; correct parse error text (#1463)

* fix: strip query/fragment from repo URLs; correct parse error text

Co-authored-by: Cursor <cursoragent@cursor.com>

* ci: pin docs deploy to Python 3.12

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix: pin ghapi<2 so sync GhApi calls do not return coroutines

Co-authored-by: Cursor <cursoragent@cursor.com>

---------

Co-authored-by: Cursor <cursoragent@cursor.com>

.github/workflows/build-docs.yaml

@@ -29,7 +29,7 @@ jobs:
           git config user.email 41898282+github-actions[bot]@users.noreply.github.com
       - uses: actions/setup-python@v6
         with:
-          python-version: 3.x
+          python-version: "3.12"
       - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV
       - uses: actions/cache@v6
         with:

pyproject.toml

@@ -56,7 +56,7 @@ dependencies = [
     "pydantic_settings",
     "litellm>=1.44.12,!=1.82.7,!=1.82.8",
     "GitPython",
-    "ghapi",
+    "ghapi>=1.0.0,<2",
     "swe-rex>=1.4.0",
     "tabulate",
     "textual>=1.0.0",

sweagent/utils/github.py

@@ -80,10 +80,10 @@ def _parse_gh_repo_url(repo_url: str) -> tuple[str, str]:
     """
     match = GITHUB_REPO_URL_PATTERN.search(repo_url)
     if not match:
-        msg = f"Invalid GitHub issue URL: {repo_url}"
+        msg = f"Invalid GitHub repository URL: {repo_url}"
         raise InvalidGithubURL(msg)
     owner, repo = match.groups()
-    return owner, repo.removesuffix(".git")
+    return owner, repo.split("?", 1)[0].split("#", 1)[0].removesuffix(".git")
 
 
 def _get_gh_issue_data(issue_url: str, *, token: str = ""):

tests/test_env_utils.py

@@ -41,10 +41,13 @@ def test_parse_gh_repo_url():
     assert _parse_gh_repo_url("git@github.com/SWE-agent/SWE-agent/asdfjsdfg") == ("SWE-agent", "SWE-agent")
     assert _parse_gh_repo_url("https://github.com/SWE-agent/SWE-agent.git") == ("SWE-agent", "SWE-agent")
     assert _parse_gh_repo_url("github.com/SWE-agent/SWE-agent.git") == ("SWE-agent", "SWE-agent")
+    assert _parse_gh_repo_url("https://github.com/SWE-agent/SWE-agent?ref=main") == ("SWE-agent", "SWE-agent")
+    assert _parse_gh_repo_url("https://github.com/SWE-agent/SWE-agent#readme") == ("SWE-agent", "SWE-agent")
+    assert _parse_gh_repo_url("https://github.com/SWE-agent/SWE-agent.git?ref=main") == ("SWE-agent", "SWE-agent")
 
 
 def test_parse_gh_repo_url_fails():
-    with pytest.raises(InvalidGithubURL):
+    with pytest.raises(InvalidGithubURL, match="repository URL"):
         _parse_gh_repo_url("adfkj;lasdfl;kj")
     with pytest.raises(InvalidGithubURL):
         _parse_gh_repo_url("github.com/")