Meaningful error message if attempting to overwrite the df attribute (

#29)

biopandas/mol2/pandas_mol2.py

@@ -52,6 +52,14 @@ def df(self):
         """Acccesses the pandas DataFrame"""
         return self._df
 
+    @df.setter
+    def df(self, value):
+        """Assign a new value to the pandas DataFrame"""
+        raise AttributeError('Please use `PandasMol2.df_ = ... ` instead\n'
+                             'of `PandasMol2.df = ... ` if you are sure that\n'
+                             'you want to overwrite the `df` attribute.')
+        # self._df = value
+
     def _load_mol2(self, mol2_lines, mol2_code, columns):
         """Load mol2 contents into assert_raise_message instance"""
         if columns is None:

biopandas/mol2/tests/test_pandas_mol2.py

@@ -8,6 +8,7 @@
 import os
 from biopandas.mol2 import PandasMol2
 from biopandas.mol2.mol2_io import split_multimol2
+from biopandas.testutils import assert_raises
 
 this_dir = os.path.dirname(os.path.realpath(__file__))
 
@@ -55,3 +56,20 @@ def test_distance():
 
     pdmol = PandasMol2().read_mol2(data_path)
     assert round(pdmol.distance().values[0], 3) == 31.185
+
+
+def test_overwrite_df():
+    data_path = os.path.join(this_dir, 'data', '1b5e_1.mol2')
+    pdmol = PandasMol2().read_mol2(data_path)
+
+    def overwrite():
+        pdmol.df = pdmol.df[(pdmol.df['atom_type'] != 'H')]
+
+    expect = ('Please use `PandasMol2.df_ = ... `'
+              ' instead\nof `PandasMol2.df = ... `'
+              ' if you are sure that\nyou want'
+              ' to overwrite the `df` attribute.')
+
+    assert_raises(AttributeError,
+                  expect,
+                  overwrite)

biopandas/pdb/pandas_pdb.py

@@ -55,6 +55,14 @@ def df(self):
         """Acccess dictionary of pandas DataFrames for PDB record sections."""
         return self._df
 
+    @df.setter
+    def df(self, value):
+        """Assign a new value to the pandas DataFrame"""
+        raise AttributeError('Please use `PandasPdb.df_ = ... ` instead\n'
+                             'of `PandasPdb.df = ... ` if you are sure that\n'
+                             'you want to overwrite the `df` attribute.')
+        # self._df = value
+
     def read_pdb(self, path):
         """Read PDB files (unzipped or gzipped) from local drive
 

biopandas/pdb/tests/test_assign_df.py

@@ -0,0 +1,28 @@
+# BioPandas
+# Author: Sebastian Raschka <mail@sebastianraschka.com>
+# License: BSD 3 clause
+# Project Website: http://rasbt.github.io/biopandas/
+# Code Repository: https://github.com/rasbt/biopandas
+
+from biopandas.pdb import PandasPdb
+from biopandas.testutils import assert_raises
+import os
+
+
+TESTDATA_FILENAME = os.path.join(os.path.dirname(__file__), 'data', '3eiy.pdb')
+
+
+def test_overwrite_df():
+    data_path = os.path.join(os.path.dirname(__file__), 'data', '3eiy.pdb')
+    pdb = PandasPdb().read_pdb(data_path)
+
+    def overwrite():
+        pdb.df = 'bla'
+
+    expect = ('Please use `PandasPdb.df_ = ... ` instead\n'
+              'of `PandasPdb.df = ... ` if you are sure that\n'
+              'you want to overwrite the `df` attribute.')
+
+    assert_raises(AttributeError,
+                  expect,
+                  overwrite)

biopandas/testutils/__init__.py

@@ -3,3 +3,5 @@
 # License: BSD 3 clause
 # Project Website: http://rasbt.github.io/biopandas/
 # Code Repository: https://github.com/rasbt/biopandas
+
+from .testutils import assert_raises

biopandas/testutils/testutils.py

@@ -9,12 +9,39 @@
 
 def assertMultiLineEqual(first, second, preserve_newline=True, msg=None):
     """Assert that two multi-line strings are equal."""
-    assert isinstance(first, str) == True, 'First argument is not a string'
-    assert isinstance(second, str) == True, 'Second argument is not a string'
+    assert isinstance(first, str), 'First argument is not a string'
+    assert isinstance(second, str), 'Second argument is not a string'
 
     if first != second:
         message = ''.join(difflib.ndiff(first.splitlines(preserve_newline),
                                         second.splitlines(preserve_newline)))
         if msg:
             message += " : " + msg
         raise AssertionError("Multi-line strings are unequal:\n" + message)
+
+
+def assert_raises(exception_type, message, func, *args, **kwargs):
+    """Check that an exception is raised with a specific message
+
+    Parameters
+    ----------
+    exception_type : exception
+        The exception that should be raised
+    message : str (default: None)
+        The error message that should be raised. Ignored if False or None
+    func : callable
+        The function that raises the exception
+    *args : positional arguments to `func`
+    **kwargs : keyword arguments to `func`
+
+    """
+    try:
+        func(*args, **kwargs)
+    except exception_type as e:
+        error_message = str(e)
+        if message and message not in error_message:
+            raise AssertionError("Error message differs from the expected"
+                                 " string: %r. Got error message: %r" %
+                                 (message, error_message))
+    else:
+        raise AssertionError('%s not raised.' % exception_type.__name__)

docs/sources/CHANGELOG.md

@@ -18,6 +18,7 @@ The CHANGELOG for the current development version is available at
 
 - The `amino3to1` method of `biopandas.pdb.PandasPDB` objects now returns a pandas `DataFrame` instead of a pandas `Series` object. The returned data frame has two columns, `'chain_id'` and `'residue_name'`, where the former contains the chain ID of the amino acid and the latter contains the 1-letter amino acid code, respectively.
 -  Significant speed improvements of the `distance` method of both `PandasPdb` and `PandasMol2` (now about 300 percent faster than previously).
+-  Add meaningful error message if attempting to overwrite the `df` attributes of `PandasMol2` and `PandasPdb` directly.
 
 ##### Bug Fixes