Added DataError for irrecoverable data interface initialization errors (

#2041)

* Added DataError for irrecoverable data interface init errors

* Added additional interface info to DataError

* Fixed DataArray vdim error

* Added DataError for missing xarray coords

* Fixed error for GriddedInterface shape mismatch

* Updated MultiInterface error tests

holoviews/core/data/array.py

@@ -5,7 +5,7 @@
 
 import numpy as np
 
-from .interface import Interface
+from .interface import Interface, DataError
 from ..dimension import Dimension
 from ..element import Element
 from ..ndmapping import NdMapping, item_check
@@ -75,8 +75,8 @@ def validate(cls, dataset):
         ndims = len(dataset.dimensions())
         ncols = dataset.data.shape[1] if dataset.data.ndim > 1 else 1
         if ncols < ndims:
-            raise ValueError("Supplied data does not match specified "
-                             "dimensions, expected at least %s columns." % ndims)
+            raise DataError("Supplied data does not match specified "
+                            "dimensions, expected at least %s columns." % ndims, cls)
 
 
     @classmethod

holoviews/core/data/dictionary.py

@@ -8,7 +8,7 @@
 
 import numpy as np
 
-from .interface import Interface
+from .interface import Interface, DataError
 from ..dimension import Dimension
 from ..element import Element
 from ..dimension import OrderedDict as cyODict
@@ -104,10 +104,14 @@ def validate(cls, dataset):
         dimensions = dataset.dimensions(label='name')
         not_found = [d for d in dimensions if d not in dataset.data]
         if not_found:
-            raise ValueError('Following dimensions not found in data: %s' % not_found)
-        lengths = [len(dataset.data[dim]) for dim in dimensions if not np.isscalar(dataset.data[dim])]
-        if len({l for l in lengths if l > 1}) > 1:
-            raise ValueError('Length of columns do not match')
+            raise DataError('Following columns specified as dimensions '
+                            'but not found in data: %s' % not_found, cls)
+        lengths = [(dim, 1 if np.isscalar(dataset.data[dim]) else len(dataset.data[dim]))
+                   for dim in dimensions]
+        if len({l for d, l in lengths if l > 1}) > 1:
+            lengths = ', '.join(['%s: %d' % l for l in sorted(lengths)])
+            raise DataError('Length of columns must be equal or scalar, '
+                            'columns have lengths: %s' % lengths, cls)
 
 
     @classmethod

holoviews/core/data/grid.py

@@ -8,7 +8,7 @@
 import numpy as np
 
 from .dictionary import DictInterface
-from .interface import Interface
+from .interface import Interface, DataError
 from ..dimension import Dimension
 from ..element import Element
 from ..dimension import OrderedDict as cyODict
@@ -74,10 +74,11 @@ def init(cls, eltype, data, kdims, vdims):
         expected = tuple([len(data[kd]) for kd in kdim_names])
         for vdim in vdim_names:
             shape = data[vdim].shape
+            error = DataError if len(shape) > 1 else ValueError
             if shape != expected[::-1] and not (not expected and shape == (1,)):
-                raise ValueError('Key dimension values and value array %s '
-                                 'shape do not match. Expected shape %s, '
-                                 'actual shape: %s' % (vdim, expected[::-1], shape))
+                raise error('Key dimension values and value array %s '
+                            'shapes do not match. Expected shape %s, '
+                            'actual shape: %s' % (vdim, expected[::-1], shape), cls)
         return data, {'kdims':kdims, 'vdims':vdims}, {}
 
 

holoviews/core/data/interface.py

@@ -6,6 +6,15 @@
 from .. import util
 
 
+class DataError(ValueError):
+    "DataError is raised when the data cannot be interpreted"
+
+    def __init__(self, msg, interface=None):
+        if interface is not None:
+            msg = '\n\n'.join([msg, interface.error()])
+        super(DataError, self).__init__(msg)
+
+
 class iloc(object):
     """
     iloc is small wrapper object that allows row, column based
@@ -121,6 +130,24 @@ def cast(cls, dataset, datatype=None, cast_type=None):
                 for co in dataset]
 
 
+    @classmethod
+    def error(cls):
+        info = dict(interface=cls.__name__)
+        url = "http://holoviews.org/user_guide/%s_Datasets.html"
+        if cls.multi:
+            datatype = 'a list of tabular'
+            info['url'] = url % 'Tabular'
+        else:
+            if cls.gridded:
+                datatype = 'gridded'
+            else:
+                datatype = 'tabular'
+            info['url'] = url % datatype.capitalize()
+        info['datatype'] = datatype
+        return ("{interface} expects {datatype} data, for more information "
+                "on supported datatypes see {url}".format(**info))
+
+
     @classmethod
     def initialize(cls, eltype, data, kdims, vdims, datatype=None):
         # Process params and dimensions
@@ -162,6 +189,8 @@ def initialize(cls, eltype, data, kdims, vdims, datatype=None):
             try:
                 (data, dims, extra_kws) = interface.init(eltype, data, kdims, vdims)
                 break
+            except DataError:
+                raise
             except Exception:
                 pass
         else:
@@ -176,9 +205,9 @@ def validate(cls, dataset):
         not_found = [d for d in dataset.dimensions(label='name')
                      if d not in dataset.data]
         if not_found:
-            raise ValueError("Supplied data does not contain specified "
-                             "dimensions, the following dimensions were "
-                             "not found: %s" % repr(not_found))
+            raise DataError("Supplied data does not contain specified "
+                            "dimensions, the following dimensions were "
+                            "not found: %s" % repr(not_found), cls)
 
 
     @classmethod

holoviews/core/data/iris.py

@@ -8,7 +8,7 @@
 
 import numpy as np
 
-from .interface import Interface
+from .interface import Interface, DataError
 from .grid import GridInterface
 from ..dimension import Dimension
 from ..element import Element
@@ -129,7 +129,7 @@ def init(cls, eltype, data, kdims, vdims):
     @classmethod
     def validate(cls, dataset):
         if len(dataset.vdims) > 1:
-            raise ValueError("Iris cubes do not support more than one value dimension")
+            raise DataError("Iris cubes do not support more than one value dimension", cls)
 
 
     @classmethod

holoviews/core/data/multipath.py

@@ -1,7 +1,8 @@
 import numpy as np
 
 from ..util import max_range
-from .interface import Interface
+from .interface import Interface, DataError
+
 
 class MultiInterface(Interface):
     """
@@ -40,11 +41,11 @@ def init(cls, eltype, data, kdims, vdims):
                                                          datatype=cls.subtypes)
             if prev_interface:
                 if prev_interface != interface:
-                    raise ValueError('MultiInterface subpaths must all have matching datatype.')
+                    raise DataError('MultiInterface subpaths must all have matching datatype.', cls)
                 if dims['kdims'] != prev_dims['kdims']:
-                    raise ValueError('MultiInterface subpaths must all have matching kdims.')
+                    raise DataError('MultiInterface subpaths must all have matching kdims.', cls)
                 if dims['vdims'] != prev_dims['vdims']:
-                    raise ValueError('MultiInterface subpaths must all have matching vdims.')
+                    raise DataError('MultiInterface subpaths must all have matching vdims.', cls)
             new_data.append(d)
             prev_interface, prev_dims = interface, dims
         return new_data, dims, {}

holoviews/core/data/pandas.py

@@ -10,7 +10,7 @@
 import numpy as np
 import pandas as pd
 
-from .interface import Interface
+from .interface import Interface, DataError
 from ..dimension import Dimension
 from ..element import Element
 from ..dimension import OrderedDict as cyODict
@@ -44,6 +44,9 @@ def init(cls, eltype, data, kdims, vdims):
             elif kdims is None and vdims is None:
                 kdims = list(data.columns[:ndim])
                 vdims = [] if ndim is None else list(data.columns[ndim:])
+            if any(isinstance(d, (np.int64, int)) for d in kdims+vdims):
+                raise DataError("pandas DataFrame column names used as dimensions "
+                                "must be strings not integers.", cls)
         else:
             # Check if data is of non-numeric type
             # Then use defined data type
@@ -92,9 +95,9 @@ def validate(cls, dataset):
         not_found = [d for d in dataset.dimensions(label='name')
                      if d not in dataset.data.columns]
         if not_found:
-            raise ValueError("Supplied data does not contain specified "
-                             "dimensions, the following dimensions were "
-                             "not found: %s" % repr(not_found))
+            raise DataError("Supplied data does not contain specified "
+                            "dimensions, the following dimensions were "
+                            "not found: %s" % repr(not_found), cls)
 
 
     @classmethod

holoviews/core/data/xarray.py

@@ -16,7 +16,7 @@
 from ..ndmapping import NdMapping, item_check, sorted_context
 from ..element import Element
 from .grid import GridInterface
-from .interface import Interface
+from .interface import Interface, DataError
 
 
 class XArrayInterface(GridInterface):
@@ -50,6 +50,9 @@ def init(cls, eltype, data, kdims, vdims):
                 vdim = vdims[0]
             elif len(vdim_param.default) == 1:
                 vdim = vdim_param.default[0]
+            else:
+                raise DataError("If xarray DataArray does not define a name "
+                                "an explicit vdim must be supplied.", cls)
             vdims = [vdim]
             if not kdims:
                 kdims = [Dimension(d) for d in data.dims[::-1]]
@@ -88,8 +91,13 @@ def init(cls, eltype, data, kdims, vdims):
                 kdims = [name for name in data.indexes.keys()
                          if isinstance(data[name].data, np.ndarray)]
 
+        not_found = [d for d in kdims if d.name not in data.coords]
         if not isinstance(data, xr.Dataset):
             raise TypeError('Data must be be an xarray Dataset type.')
+        elif not_found:
+            raise DataError("xarray Dataset must define coordinates "
+                            "for all defined kdims, %s coordinates not found."
+                            % not_found, cls)
         return data, {'kdims': kdims, 'vdims': vdims}, {}
 
 

tests/testmultiinterface.py

@@ -6,6 +6,7 @@
 
 import numpy as np
 from holoviews import Dataset
+from holoviews.core.data.interface import DataError
 from holoviews.element import Path
 from holoviews.element.comparison import ComparisonTestCase
 
@@ -112,15 +113,13 @@ def test_multi_array_redim(self):
     def test_multi_mixed_interface_raises(self):
         arrays = [np.random.rand(10, 2) if j else {'x': range(10), 'y': range(10)}
                   for i in range(2) for j in range(2)]
-        error = "None of the available storage backends were able to support the supplied data format."
-        with self.assertRaisesRegexp(ValueError, error):
+        with self.assertRaises(DataError):
             mds = Path(arrays, kdims=['x', 'y'], datatype=['multitabular'])
 
     def test_multi_mixed_dims_raises(self):
         arrays = [{'x': range(10), 'y' if j else 'z': range(10)}
                   for i in range(2) for j in range(2)]
-        error = "Following dimensions not found in data: \['y'\]"
-        with self.assertRaisesRegexp(ValueError, error):
+        with self.assertRaises(DataError):
             mds = Path(arrays, kdims=['x', 'y'], datatype=['multitabular'])
 
     def test_multi_split(self):