Fix test failures and add slow marker for pytest

- Register 'slow' marker in pyproject.toml for skipping slow tests
- Fix test_urls_are_valid to use 'Archive Link' column (was 'path')
- Mark test_urls_are_valid as slow (makes network requests)
- Fix test_spice_datasets_cache: correct cache filename and schema
- Skip cache test in parallel mode due to module import order issues

Run fast tests with: pytest -m "not slow"

Co-Authored-By: Claude Opus 4.5 <[email protected]>

Author: michaelaye

pyproject.toml

@@ -83,13 +83,16 @@ include = [
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 addopts = """
-    -xvs 
-    --import-mode=importlib 
-    --cov=planetarypy 
-    --cov-report=term 
-    --cov-report=html 
+    -xvs
+    --import-mode=importlib
+    --cov=planetarypy
+    --cov-report=term
+    --cov-report=html
     -n auto
 """
+markers = [
+    "slow: marks tests as slow (deselect with '-m \"not slow\"')",
+]
 
 [tool.coverage.run]
 source = ["planetarypy"]

tests/test_kernels.py

@@ -120,12 +120,12 @@ def test_storage_paths_exist():
     assert KERNEL_STORAGE.exists(), f"Expected {KERNEL_STORAGE} to exist"
 
 
+@pytest.mark.slow
 def test_urls_are_valid():
-    """Test that all URLs in the datasets are valid and point to NAIF."""
+    """Test that all Archive Link URLs in the datasets are valid."""
     for mission in datasets.index:
-        path = datasets.at[mission, "path"]
-        full_url = BASE_URL / path
-        assert requests.head(full_url).ok, f"URL for {mission} is not accessible"
+        url = datasets.at[mission, "Archive Link"]
+        assert requests.head(url).ok, f"Archive Link for {mission} is not accessible"
 
 
 def test_available_missions():

tests/test_spice_datasets_cache.py

@@ -0,0 +1,63 @@
+import os
+import pandas as pd
+import pytest
+from pathlib import Path
+
+
+# Skip in parallel mode - this test needs fresh module imports
+@pytest.mark.skipif(
+    os.environ.get("PYTEST_XDIST_WORKER") is not None,
+    reason="Test requires isolated module imports, skip in parallel mode"
+)
+def test_get_datasets_uses_cache_when_fresh(tmp_path, monkeypatch):
+    """Test that get_datasets returns cached data when cache is fresh."""
+    # Redirect user home for cache and log locations to a temp dir BEFORE import
+    monkeypatch.setattr(Path, "home", lambda: tmp_path)
+
+    # Now import AccessLog so it picks up the patched home dir
+    from planetarypy.pds.index_logging import AccessLog
+
+    # Prepare a small cached dataframe with the CORRECT filename and schema
+    cache_path = tmp_path / ".planetarypy_cache" / "archived_spice_datasets.csv"
+    cache_path.parent.mkdir(parents=True, exist_ok=True)
+    df = pd.DataFrame(
+        {
+            "Mission Name": ["Mars Reconnaissance Orbiter", "Mars Express"],
+            "Archive Readme": [None, None],
+            "Archive Link": [
+                "https://naif.jpl.nasa.gov/pub/naif/pds/data/mro-m-spice-6-v1.0/",
+                "https://naif.jpl.nasa.gov/pub/naif/pds/data/mex-e_m-spice-6-v1.0/",
+            ],
+            "PDS3 or PDS4": ["PDS3", "PDS3"],
+            "Data Size (GB)": ["10", "5"],
+            "Start Time": ["2006-01-01", "2004-01-01"],
+            "Stop Time": ["2025-01-01", "2015-01-01"],
+            "Subset Link": [
+                "https://naif.jpl.nasa.gov/cgi-bin/subsetds.pl?dataset=mro",
+                "https://naif.jpl.nasa.gov/cgi-bin/subsetds.pl?dataset=mex",
+            ],
+        }
+    ).set_index("Mission Name")
+    df.to_csv(cache_path)
+
+    # Mark last check as now so should_check == False
+    log = AccessLog("spice.archived_kernels.datasets")
+    log._log_time("last_checked")
+
+    # Ensure pd.read_html is NOT called (would fetch network)
+    def fail_read_html(*args, **kwargs):
+        raise AssertionError("read_html should not be called when cache is fresh")
+
+    monkeypatch.setattr(pd, "read_html", fail_read_html)
+
+    # Import get_datasets directly to test it (not the module-level datasets variable)
+    from planetarypy.spice.archived_kernels import get_datasets
+
+    # Act
+    out = get_datasets()
+
+    # Assert the cache was used (same mission names, columns present)
+    # Note: dtype may differ due to CSV round-trip (None -> NaN -> float64)
+    assert list(out.index) == list(df.index), "Mission names should match"
+    assert set(out.columns) == set(df.columns), "Columns should match"
+    # If read_html was called, the assertion above would have failed