From 17c9e8fa50da610990c42f0a49d63754a9a7db0e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Kai=20M=C3=BChlbauer?= <kmuehlbauer@wradlib.org>
Date: Thu, 6 Jul 2023 17:11:49 +0200
Subject: [PATCH] ensure no forward slashes in names for HDF5-based backends
 (#7953)

* ensure no forward slashes in names for HDF5-based backends

* fix mypy

* Update xarray/backends/netCDF4_.py

Co-authored-by: Joe Hamman <jhamman1@gmail.com>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: Joe Hamman <jhamman1@gmail.com>
Co-authored-by: Deepak Cherian <dcherian@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 doc/whats-new.rst             |  3 +++
 xarray/backends/h5netcdf_.py  |  3 +++
 xarray/backends/netCDF4_.py   | 12 ++++++++++++
 xarray/tests/test_backends.py | 14 ++++++++++++++
 4 files changed, 32 insertions(+)

diff --git a/doc/whats-new.rst b/doc/whats-new.rst
index be7f8ade57c..87559ee9b02 100644
--- a/doc/whats-new.rst
+++ b/doc/whats-new.rst
@@ -34,6 +34,9 @@ Deprecations
 Bug fixes
 ~~~~~~~~~
 
+- Ensure no forward slashes in variable and dimension names for HDF5-based engines.
+  (:issue:`7943`, :pull:`7953`) By `Kai Mühlbauer <https://github.com/kmuehlbauer>`_.
+
 
 Documentation
 ~~~~~~~~~~~~~
diff --git a/xarray/backends/h5netcdf_.py b/xarray/backends/h5netcdf_.py
index 7389f6a2862..697ebb8ab92 100644
--- a/xarray/backends/h5netcdf_.py
+++ b/xarray/backends/h5netcdf_.py
@@ -20,6 +20,7 @@
 from xarray.backends.netCDF4_ import (
     BaseNetCDF4Array,
     _encode_nc4_variable,
+    _ensure_no_forward_slash_in_name,
     _extract_nc4_variable_encoding,
     _get_datatype,
     _nc4_require_group,
@@ -256,6 +257,7 @@ def get_encoding(self):
             }
 
     def set_dimension(self, name, length, is_unlimited=False):
+        _ensure_no_forward_slash_in_name(name)
         if is_unlimited:
             self.ds.dimensions[name] = None
             self.ds.resize_dimension(name, length)
@@ -273,6 +275,7 @@ def prepare_variable(
     ):
         import h5py
 
+        _ensure_no_forward_slash_in_name(name)
         attrs = variable.attrs.copy()
         dtype = _get_datatype(variable, raise_on_invalid_encoding=check_encoding)
 
diff --git a/xarray/backends/netCDF4_.py b/xarray/backends/netCDF4_.py
index 8a5d48c8c1e..b5c3413e7f8 100644
--- a/xarray/backends/netCDF4_.py
+++ b/xarray/backends/netCDF4_.py
@@ -194,6 +194,15 @@ def _nc4_require_group(ds, group, mode, create_group=_netcdf4_create_group):
         return ds
 
 
+def _ensure_no_forward_slash_in_name(name):
+    if "/" in name:
+        raise ValueError(
+            f"Forward slashes '/' are not allowed in variable and dimension names (got {name!r}). "
+            "Forward slashes are used as hierarchy-separators for "
+            "HDF5-based files ('netcdf4'/'h5netcdf')."
+        )
+
+
 def _ensure_fill_value_valid(data, attributes):
     # work around for netCDF4/scipy issue where _FillValue has the wrong type:
     # https://github.com/Unidata/netcdf4-python/issues/271
@@ -447,6 +456,7 @@ def get_encoding(self):
         }
 
     def set_dimension(self, name, length, is_unlimited=False):
+        _ensure_no_forward_slash_in_name(name)
         dim_length = length if not is_unlimited else None
         self.ds.createDimension(name, size=dim_length)
 
@@ -470,6 +480,8 @@ def encode_variable(self, variable):
     def prepare_variable(
         self, name, variable, check_encoding=False, unlimited_dims=None
     ):
+        _ensure_no_forward_slash_in_name(name)
+
         datatype = _get_datatype(
             variable, self.format, raise_on_invalid_encoding=check_encoding
         )
diff --git a/xarray/tests/test_backends.py b/xarray/tests/test_backends.py
index 0450e769e7b..dad2d668ff8 100644
--- a/xarray/tests/test_backends.py
+++ b/xarray/tests/test_backends.py
@@ -1611,6 +1611,20 @@ def test_encoding_unlimited_dims(self) -> None:
             assert actual.encoding["unlimited_dims"] == set("y")
             assert_equal(ds, actual)
 
+    def test_raise_on_forward_slashes_in_names(self) -> None:
+        # test for forward slash in variable names and dimensions
+        # see GH 7943
+        data_vars: list[dict[str, Any]] = [
+            {"PASS/FAIL": (["PASSFAIL"], np.array([0]))},
+            {"PASS/FAIL": np.array([0])},
+            {"PASSFAIL": (["PASS/FAIL"], np.array([0]))},
+        ]
+        for dv in data_vars:
+            ds = Dataset(data_vars=dv)
+            with pytest.raises(ValueError, match="Forward slashes '/' are not allowed"):
+                with self.roundtrip(ds):
+                    pass
+
 
 @requires_netCDF4
 class TestNetCDF4Data(NetCDF4Base):