ENH/API: allow DataFrame constructor to better accept list-like collections (GH3783,GH4297)

doc/source/release.rst

@@ -113,6 +113,8 @@ Improvements to existing features
     ``io.excel.xls.writer``. (:issue:`4745`, :issue:`4750`)
   - ``Panel.to_excel()`` now accepts keyword arguments that will be passed to
     its ``DataFrame``'s ``to_excel()`` methods. (:issue:`4750`)
+  - allow DataFrame constructor to accept more list-like objects, e.g. list of
+    ``collections.Sequence`` and ``array.Array`` objects (:issue:`3783`,:issue:`42971`)
 
 API Changes
 ~~~~~~~~~~~

pandas/core/frame.py

@@ -16,6 +16,7 @@
 import sys
 import collections
 import warnings
+import types
 
 from numpy import nan as NA
 import numpy as np
@@ -24,7 +25,7 @@
 from pandas.core.common import (isnull, notnull, PandasError, _try_sort,
                                 _default_index, _maybe_upcast, _is_sequence,
                                 _infer_dtype_from_scalar, _values_from_object,
-                                _coerce_to_dtypes, _DATELIKE_DTYPES)
+                                _coerce_to_dtypes, _DATELIKE_DTYPES, is_list_like)
 from pandas.core.generic import NDFrame
 from pandas.core.index import Index, MultiIndex, _ensure_index
 from pandas.core.indexing import (_NDFrameIndexer, _maybe_droplevels,
@@ -413,12 +414,14 @@ def __init__(self, data=None, index=None, columns=None, dtype=None,
             else:
                 mgr = self._init_ndarray(data, index, columns, dtype=dtype,
                                          copy=copy)
-        elif isinstance(data, list):
+        elif isinstance(data, (list, types.GeneratorType)):
+            if isinstance(data, types.GeneratorType):
+                data = list(data)
             if len(data) > 0:
                 if index is None and isinstance(data[0], Series):
                     index = _get_names_from_index(data)
 
-                if isinstance(data[0], (list, tuple, collections.Mapping, Series)):
+                if is_list_like(data[0]) and getattr(data[0],'ndim',0) <= 1:
                     arrays, columns = _to_arrays(data, columns, dtype=dtype)
                     columns = _ensure_index(columns)
 
@@ -4545,7 +4548,7 @@ def isin(self, values, iloc=False):
 
 
         else:
-            if not com.is_list_like(values):
+            if not is_list_like(values):
                 raise TypeError("only list-like or dict-like objects are"
                                 " allowed to be passed to DataFrame.isin(), "
                                 "you passed a "
@@ -4705,7 +4708,7 @@ def extract_index(data):
             elif isinstance(v, dict):
                 have_dicts = True
                 indexes.append(list(v.keys()))
-            elif isinstance(v, (list, tuple, np.ndarray)):
+            elif is_list_like(v) and getattr(v,'ndim',0) <= 1:
                 have_raw_arrays = True
                 raw_lengths.append(len(v))
 

pandas/tests/test_frame.py

@@ -2606,6 +2606,57 @@ def test_constructor_list_of_lists(self):
         self.assert_(com.is_integer_dtype(df['num']))
         self.assert_(df['str'].dtype == np.object_)
 
+    def test_constructor_sequence_like(self):
+        # GH 3783
+        # collections.Squence like
+        import collections
+
+        class DummyContainer(collections.Sequence):
+            def __init__(self, lst):
+                self._lst = lst
+            def __getitem__(self, n):
+                return self._lst.__getitem__(n)
+            def __len__(self, n):
+                return self._lst.__len__()
+
+        l = [DummyContainer([1, 'a']), DummyContainer([2, 'b'])]
+        columns = ["num", "str"]
+        result = DataFrame(l, columns=columns)
+        expected = DataFrame([[1,'a'],[2,'b']],columns=columns)
+        assert_frame_equal(result, expected, check_dtype=False)
+
+        # GH 4297
+        # support Array
+        import array
+        result = DataFrame.from_items([('A', array.array('i', range(10)))])
+        expected = DataFrame({ 'A' : list(range(10)) })
+        assert_frame_equal(result, expected, check_dtype=False)
+
+        expected = DataFrame([ list(range(10)), list(range(10)) ])
+        result = DataFrame([ array.array('i', range(10)), array.array('i',range(10)) ])
+        assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_constructor_iterator(self):
+
+        expected = DataFrame([ list(range(10)), list(range(10)) ])
+        result = DataFrame([ range(10), range(10) ])
+        assert_frame_equal(result, expected)
+
+    def test_constructor_generator(self):
+        #related #2305
+
+        gen1 = (i for i in range(10))
+        gen2 = (i for i in range(10))
+
+        expected = DataFrame([ list(range(10)), list(range(10)) ])
+        result = DataFrame([ gen1, gen2 ])
+        assert_frame_equal(result, expected)
+
+        gen = ([ i, 'a'] for i in range(10))
+        result = DataFrame(gen)
+        expected = DataFrame({ 0 : range(10), 1 : 'a' })
+        assert_frame_equal(result, expected)
+
     def test_constructor_list_of_dicts(self):
         data = [OrderedDict([['a', 1.5], ['b', 3], ['c', 4], ['d', 6]]),
                 OrderedDict([['a', 1.5], ['b', 3], ['d', 6]]),

pandas/tests/test_series.py

@@ -353,6 +353,12 @@ def test_constructor_series(self):
 
         assert_series_equal(s2, s1.sort_index())
 
+    def test_constructor_iterator(self):
+
+        expected = Series(list(range(10)))
+        result = Series(range(10))
+        assert_series_equal(result, expected)
+
     def test_constructor_generator(self):
         gen = (i for i in range(10))