Silence warnings about writing NPZ files with StringDType

Author: otsaloma

NEWS.md

@@ -1,3 +1,11 @@
+PENDING: Dataiter 1.0
+=====================
+
+* Silence warnings about writing NPZ files with StringDType:
+  "UserWarning: Custom dtypes are saved as python objects using the
+  pickle protocol. Loading this file requires allow_pickle=True to be
+  set."
+
 2025-01-12: Dataiter 0.9999
 ===========================
 

dataiter/dtypes.py

@@ -20,6 +20,8 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 # THE SOFTWARE.
 
+import warnings
+
 from numpy.dtypes import StringDType
 
 #: Instance of NumPy variable-width StringDType used
@@ -30,3 +32,7 @@
 # input, a distinction between NA and blank cannot usually be made.
 # TODO: Consider changing this to something like ':NA:'.
 # https://numpy.org/doc/stable/user/basics.strings.html#missing-data-support
+
+# Ignore pointless warnings about using StringDType in numpy.savez.
+_pattern = "Custom dtypes are saved as python objects using the pickle protocol"
+warnings.filterwarnings("ignore", message=_pattern, category=UserWarning)

dataiter/test/test_data_frame.py

@@ -31,9 +31,8 @@
 from pathlib import Path
 from unittest.mock import patch
 
-# This UserWarning seems unavoidable when using StringDtype, which unlike
-# other dtypes must be instantiated and that custom instance used.
-IGNORE_NPZ_PICKLE_WARNING = "ignore:Custom dtypes are saved as python objects using the pickle protocol."
+# See dataiter/dtypes.py for the equivalent non-test ignore.
+IGNORE_NPZ_PICKLE_WARNING = "ignore:Custom dtypes are saved as python objects using the pickle protocol"
 
 class TestDataFrameColumn: