Validate downloaded file integrity and raise ValueError on hash mismatch

Signed-off-by: Enoch Mok <enochmokny@gmail.com>

Author: e-mny

monai/apps/utils.py

@@ -268,9 +268,11 @@ def download_url(
         pass
     logger.info(f"Downloaded: {filepath}")
     if not check_hash(filepath, hash_val, hash_type):
-        raise RuntimeError(
-            f"{hash_type} check of downloaded file failed: URL={url}, "
-            f"filepath={filepath}, expected {hash_type}={hash_val}."
+        raise ValueError(
+            f"{hash_type} hash check of downloaded file failed: URL={url}, "
+            f"filepath={filepath}, expected {hash_type}={hash_val}, "
+            f"The file may be corrupted or tampered with. "
+            "Please retry the download or verify the source."
         )
 
 

tests/test_utils.py

@@ -79,7 +79,7 @@
     "unexpected EOF",  # incomplete download
     "network issue",
     "gdown dependency",  # gdown not installed
-    "md5 check",
+    "hash check",    # check hash value of downloaded file
     "limit",  # HTTP Error 503: Egress is over the account limit
     "authenticate",
     "timed out",  # urlopen error [Errno 110] Connection timed out
@@ -180,8 +180,11 @@ def skip_if_downloading_fails():
         err_str = str(rt_e)
         if any(k in err_str for k in DOWNLOAD_FAIL_MSGS):
             raise unittest.SkipTest(f"Error while downloading: {rt_e}") from rt_e  # incomplete download
-
+        
         raise rt_e
+    except ValueError as v_e:
+        if "hash check" in str(v_e):
+            raise unittest.SkipTest(f"Hash value error while downloading: {v_e}") from v_e
 
 
 SAMPLE_TIFF = "https://huggingface.co/datasets/MONAI/testing_data/resolve/main/CMU-1.tiff"