scikit-learn · rth · Mar 18, 2021 · Mar 13, 2021 · Mar 13, 2021
diff --git a/doc/whats_new/v1.0.rst b/doc/whats_new/v1.0.rst
@@ -68,6 +68,9 @@ Changelog
   `Thomas Fan`_ and :user:`Amanda Dsouza <amy12xx>` and
   :user:`EL-ATEIF Sara <elateifsara>`.
 
+- |Enhancement| :func:`datasets.fetch_kddcup99` raises a better message
+  when the cached file is invalid. :pr:`19669` `Thomas Fan`_.
+
 :mod:`sklearn.decomposition`
 ............................
 

diff --git a/sklearn/conftest.py b/sklearn/conftest.py
@@ -35,8 +35,11 @@ def wrapped(*args, **kwargs):
         kwargs['download_if_missing'] = download_if_missing
         try:
             return f(*args, **kwargs)
-        except IOError:
-            pytest.skip("test is enabled when SKLEARN_SKIP_NETWORK_TESTS=0")
+        except IOError as e:
+            if str(e) != "Data not found and `download_if_missing` is False":
+                raise
+            pytest.skip("test is enabled when "
+                        "SKLEARN_SKIP_NETWORK_TESTS=0")
     return pytest.fixture(lambda: wrapped)
 
 

diff --git a/sklearn/datasets/_kddcup99.py b/sklearn/datasets/_kddcup99.py
@@ -315,7 +315,17 @@ def _fetch_brute_kddcup99(data_home=None,
     column_names = [c[0] for c in dt]
     target_names = column_names[-1]
     feature_names = column_names[:-1]
-    if download_if_missing and not available:
+
+    if available:
+        try:
+            X = joblib.load(samples_path)
+            y = joblib.load(targets_path)
+        except Exception as e:
+            raise IOError(
+                "The cache for fetch_kddcup99 is invalid, please delete "
+                f"{str(kddcup_dir)} and run the fetch_kddcup99 again") from e
+
+    elif download_if_missing:
         _mkdirp(kddcup_dir)
         logger.info("Downloading %s" % archive.url)
         _fetch_remote(archive, dirname=kddcup_dir)
@@ -343,15 +353,8 @@ def _fetch_brute_kddcup99(data_home=None,
 
         joblib.dump(X, samples_path, compress=0)
         joblib.dump(y, targets_path, compress=0)
-    elif not available:
-        if not download_if_missing:
-            raise IOError("Data not found and `download_if_missing` is False")
-
-    try:
-        X, y
-    except NameError:
-        X = joblib.load(samples_path)
-        y = joblib.load(targets_path)
+    else:
+        raise IOError("Data not found and `download_if_missing` is False")
 
     return Bunch(
         data=X,

diff --git a/sklearn/datasets/tests/test_kddcup99.py b/sklearn/datasets/tests/test_kddcup99.py
@@ -58,3 +58,19 @@ def test_fetch_kddcup99_shuffle(fetch_kddcup99_fxt):
 
 def test_pandas_dependency_message(fetch_kddcup99_fxt, hide_available_pandas):
     check_pandas_dependency_message(fetch_kddcup99_fxt)
+
+
+def test_corrupted_file_error_message(fetch_kddcup99_fxt, tmp_path):
+    """Check that a nice error message is raised when cache is corrupted."""
+    kddcup99_dir = tmp_path / "kddcup99_10-py3"
+    kddcup99_dir.mkdir()
+    samples_path = kddcup99_dir / "samples"
+
+    with samples_path.open("wb") as f:
+        f.write(b"THIS IS CORRUPTED")
+
+    msg = (f"The cache for fetch_kddcup99 is invalid, please "
+           f"delete {str(kddcup99_dir)} and run the fetch_kddcup99 again")
+
+    with pytest.raises(IOError, match=msg):
+        fetch_kddcup99_fxt(data_home=str(tmp_path))