TestURLDetection Class — langchain Architecture

Source Code

libs/langchain_v1/tests/unit_tests/agents/middleware/implementations/test_pii.py lines 175–223

class TestURLDetection:
    """Test URL detection."""

    def test_detect_http_url(self) -> None:
        content = "Visit http://example.com for details."
        matches = detect_url(content)

        assert len(matches) == 1
        assert matches[0]["type"] == "url"
        assert matches[0]["value"] == "http://example.com"

    def test_detect_https_url(self) -> None:
        content = "Visit https://secure.example.com/path"
        matches = detect_url(content)

        assert len(matches) == 1
        assert matches[0]["value"] == "https://secure.example.com/path"

    def test_detect_www_url(self) -> None:
        content = "Check www.example.com"
        matches = detect_url(content)

        assert len(matches) == 1
        assert matches[0]["value"] == "www.example.com"

    def test_detect_bare_domain_with_path(self) -> None:
        content = "Go to example.com/page"
        matches = detect_url(content)

        assert len(matches) == 1
        assert matches[0]["value"] == "example.com/page"

    def test_detect_multiple_urls(self) -> None:
        content = "Visit http://test.com and https://example.org"
        matches = detect_url(content)

        assert len(matches) == 2

    def test_no_url(self) -> None:
        content = "No URLs here."
        matches = detect_url(content)
        assert len(matches) == 0

    def test_bare_domain_without_path_not_detected(self) -> None:
        # To reduce false positives, bare domains without paths are not detected
        content = "The word example.com in prose"
        detect_url(content)
        # May or may not detect depending on implementation
        # This is acceptable