adding tests for corrupted files

tests/core/data/garbage.csv

@@ -0,0 +1,2 @@
+a,b,c
+1,2,"3

tests/core/test_readers.py

@@ -4,6 +4,7 @@
 from logging import CRITICAL, DEBUG, INFO
 
 import pytest
+from pandas.errors import ParserError
 from xarray import DataArray
 
 from tests.conftest import log_check
@@ -93,7 +94,11 @@ def test_func():
     ],
 )
 def test_load_netcdf(shared_datadir, caplog, file, file_var, exp_err, expected_log):
-    """Test the netdcf variable loader."""
+    """Test the netdcf variable loader.
+
+    The FileNotFoundError test is dependent on running the test_file_format_loader tests
+    first, and will fail if run independently. See TODO above.
+    """
 
     from virtual_ecosystem.core.readers import load_netcdf
 
@@ -105,6 +110,70 @@ def test_load_netcdf(shared_datadir, caplog, file, file_var, exp_err, expected_l
     log_check(caplog, expected_log)
 
 
+@pytest.mark.parametrize(
+    argnames=["file", "file_var", "exp_err", "expected_log"],
+    argvalues=[
+        (
+            "not_there.csv",
+            "irrelevant",
+            pytest.raises(FileNotFoundError),
+            ((CRITICAL, "Data file not found"),),
+        ),
+        (
+            "garbage.csv",
+            "irrelevant",
+            pytest.raises(ParserError),
+            ((CRITICAL, "Could not load data from"),),
+        ),
+        (
+            "reader_test.csv",
+            "missing",
+            pytest.raises(KeyError),
+            ((CRITICAL, "Variable missing not found in"),),
+        ),
+        (
+            "reader_test.csv",
+            "var1",
+            does_not_raise(),
+            (),
+        ),
+        (
+            "garbage.xlsx",
+            "irrelevant",
+            pytest.raises(Exception),
+            ((CRITICAL, "Unidentified exception opening"),),
+        ),
+        (
+            "reader_test.xlsx",
+            "missing",
+            pytest.raises(KeyError),
+            ((CRITICAL, "Variable missing not found in"),),
+        ),
+        (
+            "reader_test.xlsx",
+            "var1",
+            does_not_raise(),
+            (),
+        ),
+    ],
+)
+def test_load_dataframe(shared_datadir, caplog, file, file_var, exp_err, expected_log):
+    """Test the netdcf variable loader.
+
+    The FileNotFoundError test is dependent on running the test_file_format_loader tests
+    first, and will fail if run independently. See TODO above.
+    """
+
+    from virtual_ecosystem.core.readers import load_from_dataframe
+
+    with exp_err:
+        darray = load_from_dataframe(shared_datadir / file, file_var)
+        assert isinstance(darray, DataArray)
+
+    # Check the error reports
+    log_check(caplog, expected_log)
+
+
 @pytest.mark.parametrize(
     argnames=[
         "filename",

virtual_ecosystem/core/readers.py

@@ -35,6 +35,7 @@ def new_function_to_load_tif_data(...):
 from pathlib import Path
 
 from pandas import read_csv, read_excel
+from pandas.errors import ParserError
 from xarray import DataArray, load_dataset
 
 from virtual_ecosystem.core.logger import LOGGER
@@ -150,7 +151,11 @@ def load_from_dataframe(file: Path, var_name: str) -> DataArray:
 
     Raises:
         FileNotFoundError: with bad file path names.
-        ValueError: if the file data is not readable.
+        ParserError: if the csv data is not readable.
+        Exception: if the excel data is not readable.
+
+    Note: the general exception is used because of the variety of exceptions that are
+    possible with read_excel.
     """
 
     to_raise: Exception
@@ -161,13 +166,17 @@ def load_from_dataframe(file: Path, var_name: str) -> DataArray:
         if file_type == ".csv":
             dataset = read_csv(file)
         else:
-            dataset = read_excel(file)
+            dataset = read_excel(file, engine="openpyxl")
     except FileNotFoundError:
         to_raise = FileNotFoundError(f"Data file not found: {file}")
         LOGGER.critical(to_raise)
         raise to_raise
-    except ValueError as err:
-        to_raise = ValueError(f"Could not load data from {file}: {err}")
+    except ParserError as err:
+        to_raise = ParserError(f"Could not load data from {file}: {err}.")
+        LOGGER.critical(to_raise)
+        raise to_raise
+    except Exception as err:
+        to_raise = Exception(f"Unidentified exception opening {file}: {err}")
         LOGGER.critical(to_raise)
         raise to_raise
 
@@ -177,7 +186,7 @@ def load_from_dataframe(file: Path, var_name: str) -> DataArray:
         LOGGER.critical(to_raise)
         raise to_raise
 
-    return dataset[var_name]
+    return dataset[var_name].to_xarray()
 
 
 def load_to_dataarray(