WIP for MyPy CI Integration

.gitignore

@@ -61,6 +61,7 @@ dist
 .coverage
 coverage.xml
 coverage_html_report
+*.mypy_cache
 *.pytest_cache
 # hypothesis test database
 .hypothesis/

mypy.ini

@@ -0,0 +1,6 @@
+[mypy]
+ignore_missing_imports=True
+follow_imports=silent
+
+[mypy-pandas.conftest,pandas.tests.*]
+ignore_errors=True

mypy_whitelist.txt

@@ -0,0 +1,9 @@
+pandas/core/dtypes/base.py
+pandas/core/groupby/groupby.py
+pandas/core/internals/managers.py
+pandas/core/common.py
+pandas/core/arrays/timedeltas.py
+pandas/core/arrays/datetimes.py
+pandas/core/arrays/base.py
+pandas/core/frame.py
+pandas/core/indexes/base.py

pandas/core/arrays/array_.py

@@ -1,8 +1,13 @@
+from typing import Optional, Sequence, Union
+
+import numpy as np
+
 from pandas._libs import lib, tslibs
 
+from pandas.core.arrays.base import ExtensionArray
 from pandas.core.dtypes.common import (
     is_datetime64_ns_dtype, is_extension_array_dtype, is_timedelta64_ns_dtype)
-from pandas.core.dtypes.dtypes import registry
+from pandas.core.dtypes.dtypes import ExtensionDtype, registry
 
 from pandas import compat
 

pandas/core/arrays/base.py

@@ -6,6 +6,7 @@
    without warning.
 """
 import operator
+from typing import Any, Callable, Optional, Sequence, Tuple, Union
 
 import numpy as np
 
@@ -15,6 +16,7 @@
 from pandas.util._decorators import Appender, Substitution
 
 from pandas.core.dtypes.common import is_list_like
+from pandas.core.dtypes.dtypes import ExtensionDtype
 from pandas.core.dtypes.generic import ABCIndexClass, ABCSeries
 from pandas.core.dtypes.missing import isna
 
@@ -365,7 +367,7 @@ def isna(self):
         raise AbstractMethodError(self)
 
     def _values_for_argsort(self):
-        # type: () -> ndarray
+        # type: () -> np.ndarray
         """
         Return values for sorting.
 
@@ -597,7 +599,7 @@ def searchsorted(self, value, side="left", sorter=None):
         return arr.searchsorted(value, side=side, sorter=sorter)
 
     def _values_for_factorize(self):
-        # type: () -> Tuple[ndarray, Any]
+        # type: () -> Tuple[np.ndarray, Any]
         """
         Return an array and missing value suitable for factorization.
 
@@ -622,7 +624,7 @@ def _values_for_factorize(self):
         return self.astype(object), np.nan
 
     def factorize(self, na_sentinel=-1):
-        # type: (int) -> Tuple[ndarray, ExtensionArray]
+        # type: (int) -> Tuple[np.ndarray, ExtensionArray]
         """
         Encode the extension array as an enumerated type.
 

pandas/core/arrays/datetimelike.py

@@ -2,15 +2,17 @@
 from datetime import datetime, timedelta
 import operator
 import warnings
+from typing import Union, Sequence, Tuple
 
 import numpy as np
 
 from pandas._libs import NaT, algos, iNaT, lib
+from pandas._libs.tslibs.nattype import NaTType
 from pandas._libs.tslibs.period import (
     DIFFERENT_FREQ, IncompatibleFrequency, Period)
 from pandas._libs.tslibs.timedeltas import Timedelta, delta_to_nanoseconds
 from pandas._libs.tslibs.timestamps import (
-    RoundTo, maybe_integer_op_deprecated, round_nsint64)
+    RoundTo, maybe_integer_op_deprecated, round_nsint64, Timestamp)
 import pandas.compat as compat
 from pandas.compat.numpy import function as nv
 from pandas.errors import (
@@ -350,7 +352,7 @@ def __iter__(self):
 
     @property
     def asi8(self):
-        # type: () -> ndarray
+        # type: () -> np.ndarray
         """
         Integer representation of the values.
 

pandas/core/arrays/datetimes.py

@@ -2,6 +2,7 @@
 from datetime import datetime, time, timedelta
 import textwrap
 import warnings
+from typing import Union
 
 import numpy as np
 from pytz import utc

pandas/core/arrays/integer.py

@@ -510,7 +510,7 @@ def value_counts(self, dropna=True):
         return Series(array, index=index)
 
     def _values_for_argsort(self):
-        # type: () -> ndarray
+        # type: () -> np.ndarray
         """Return values for sorting.
 
         Returns

pandas/core/arrays/period.py

@@ -1,17 +1,21 @@
 # -*- coding: utf-8 -*-
 from datetime import timedelta
 import operator
+from typing import Any, Callable, List, Optional, Sequence, Union
 
 import numpy as np
 
 from pandas._libs.tslibs import (
     NaT, frequencies as libfrequencies, iNaT, period as libperiod)
+from pandas._libs.tslibs.nattype import NaTType
 from pandas._libs.tslibs.fields import isleapyear_arr
 from pandas._libs.tslibs.period import (
     DIFFERENT_FREQ, IncompatibleFrequency, Period, get_period_field_arr,
     period_asfreq_arr)
 from pandas._libs.tslibs.timedeltas import Timedelta, delta_to_nanoseconds
 import pandas.compat as compat
+from pandas.core.arrays.base import ExtensionArray
+from pandas.core.indexes.base import Index
 from pandas.util._decorators import Appender, cache_readonly
 
 from pandas.core.dtypes.common import (
@@ -132,7 +136,7 @@ class PeriodArray(dtl.DatetimeLikeArrayMixin, dtl.DatelikeOps):
     _scalar_type = Period
 
     # Names others delegate to us
-    _other_ops = []
+    _other_ops = []     # type: List[str]
     _bool_ops = ['is_leap_year']
     _object_ops = ['start_time', 'end_time', 'freq']
     _field_ops = ['year', 'month', 'day', 'hour', 'minute', 'second',

pandas/core/arrays/sparse.py

@@ -7,12 +7,13 @@
 import operator
 import re
 import warnings
+from typing import Any, Callable, Type, Union
 
 import numpy as np
 
 from pandas._libs import index as libindex, lib
 import pandas._libs.sparse as splib
-from pandas._libs.sparse import BlockIndex, IntIndex
+from pandas._libs.sparse import BlockIndex, IntIndex, SparseIndex
 from pandas._libs.tslibs import NaT
 import pandas.compat as compat
 from pandas.compat.numpy import function as nv
@@ -79,7 +80,7 @@ class SparseDtype(ExtensionDtype):
     _metadata = ('_dtype', '_fill_value', '_is_na_fill_value')
 
     def __init__(self, dtype=np.float64, fill_value=None):
-        # type: (Union[str, np.dtype, 'ExtensionDtype', type], Any) -> None
+        # type: (Union[str, np.dtype, 'ExtensionDtype', Type], Any) -> None
         from pandas.core.dtypes.missing import na_value_for_dtype
         from pandas.core.dtypes.common import (
             pandas_dtype, is_string_dtype, is_scalar
@@ -372,7 +373,7 @@ def _subtype_with_str(self):
 
 
 def _get_fill(arr):
-    # type: (SparseArray) -> ndarray
+    # type: (SparseArray) -> np.ndarray
     """
     Create a 0-dim ndarray containing the fill value
 

pandas/core/arrays/timedeltas.py

@@ -4,6 +4,7 @@
 from datetime import timedelta
 import textwrap
 import warnings
+from typing import List
 
 import numpy as np
 
@@ -134,8 +135,8 @@ class TimedeltaArray(dtl.DatetimeLikeArrayMixin, dtl.TimelikeOps):
     _scalar_type = Timedelta
     __array_priority__ = 1000
     # define my properties & methods for delegation
-    _other_ops = []
-    _bool_ops = []
+    _other_ops = []     # type: List[str]
+    _bool_ops = []      # type: List[str]
     _object_ops = ['freq']
     _field_ops = ['days', 'seconds', 'microseconds', 'nanoseconds']
     _datetimelike_ops = _field_ops + _object_ops + _bool_ops

pandas/core/base.py

@@ -15,6 +15,7 @@
 from pandas.util._decorators import Appender, Substitution, cache_readonly
 from pandas.util._validators import validate_bool_kwarg
 
+from pandas.core.arrays.base import ExtensionArray
 from pandas.core.dtypes.common import (
     is_datetime64_ns_dtype, is_datetime64tz_dtype, is_datetimelike,
     is_extension_array_dtype, is_extension_type, is_list_like, is_object_dtype,

pandas/core/common.py

@@ -9,6 +9,7 @@
 from datetime import datetime, timedelta
 from functools import partial
 import inspect
+from typing import Any
 
 import numpy as np
 

pandas/core/dtypes/base.py

@@ -1,4 +1,6 @@
 """Extend pandas with custom array types"""
+from typing import List, Optional, Type
+
 import numpy as np
 
 from pandas.errors import AbstractMethodError
@@ -211,7 +213,7 @@ def __str__(self):
 
     @property
     def type(self):
-        # type: () -> type
+        # type: () -> Type
         """
         The scalar type for the array, e.g. ``int``
 

pandas/core/frame.py

@@ -19,6 +19,7 @@
 import sys
 import warnings
 from textwrap import dedent
+from typing import List, Set, Union
 
 import numpy as np
 import numpy.ma as ma
@@ -365,7 +366,7 @@ def _constructor(self):
     _constructor_sliced = Series
     _deprecations = NDFrame._deprecations | frozenset(
         ['get_value', 'set_value', 'from_csv', 'from_items'])
-    _accessors = set()
+    _accessors = set()  # type: Set[str]
 
     @property
     def _constructor_expanddim(self):

pandas/core/groupby/groupby.py

@@ -13,10 +13,11 @@ class providing the base-class of operations.
 from functools import partial, wraps
 import types
 import warnings
+from typing import FrozenSet, Optional, Type
 
 import numpy as np
 
-from pandas._libs import Timestamp, groupby as libgroupby
+from pandas._libs import Timestamp, groupby as libgroupby  # type: ignore
 import pandas.compat as compat
 from pandas.compat import range, set_function_name, zip
 from pandas.compat.numpy import function as nv
@@ -325,7 +326,7 @@ def _group_selection_context(groupby):
 
 class _GroupBy(PandasObject, SelectionMixin):
     _group_selection = None
-    _apply_whitelist = frozenset()
+    _apply_whitelist = frozenset()  # type: FrozenSet[str]
 
     def __init__(self, obj, keys=None, axis=0, level=None,
                  grouper=None, exclusions=None, selection=None, as_index=True,
@@ -1041,7 +1042,7 @@ def _bool_agg(self, val_test, skipna):
         """
 
         def objs_to_bool(vals):
-            # type: (np.ndarray) -> (np.ndarray, typing.Type)
+            # type: (np.ndarray) -> (np.ndarray, Type)
             if is_object_dtype(vals):
                 vals = np.array([bool(x) for x in vals])
             else:
@@ -1050,7 +1051,7 @@ def objs_to_bool(vals):
             return vals.view(np.uint8), np.bool
 
         def result_to_bool(result, inference):
-            # type: (np.ndarray, typing.Type) -> np.ndarray
+            # type: (np.ndarray, Type) -> np.ndarray
             return result.astype(inference, copy=False)
 
         return self._get_cythonized_result('group_any_all', self.grouper,
@@ -1743,7 +1744,7 @@ def quantile(self, q=0.5, interpolation='linear'):
         """
 
         def pre_processor(vals):
-            # type: (np.ndarray) -> (np.ndarray, Optional[typing.Type])
+            # type: (np.ndarray) -> (np.ndarray, Optional[Type])
             if is_object_dtype(vals):
                 raise TypeError("'quantile' cannot be performed against "
                                 "'object' dtypes!")
@@ -1758,7 +1759,7 @@ def pre_processor(vals):
             return vals, inference
 
         def post_processor(vals, inference):
-            # type: (np.ndarray, Optional[typing.Type]) -> np.ndarray
+            # type: (np.ndarray, Optional[Type]) -> np.ndarray
             if inference:
                 # Check for edge case
                 if not (is_integer_dtype(inference) and
@@ -2021,7 +2022,7 @@ def _get_cythonized_result(self, how, grouper, aggregate=False,
             Function to be applied to result of Cython function. Should accept
             an array of values as the first argument and type inferences as its
             second argument, i.e. the signature should be
-            (ndarray, typing.Type).
+            (ndarray, Type).
         **kwargs : dict
             Extra arguments to be passed back to Cython funcs
 

pandas/core/indexes/base.py

@@ -2,10 +2,11 @@
 import operator
 from textwrap import dedent
 import warnings
+from typing import Union
 
 import numpy as np
 
-from pandas._libs import (
+from pandas._libs import (  # type: ignore
     algos as libalgos, index as libindex, join as libjoin, lib)
 from pandas._libs.lib import is_datetime_array
 from pandas._libs.tslibs import OutOfBoundsDatetime, Timedelta, Timestamp

pandas/core/indexes/datetimelike.py

@@ -4,6 +4,7 @@
 """
 import operator
 import warnings
+from typing import Set
 
 import numpy as np
 
@@ -698,9 +699,9 @@ class DatetimelikeDelegateMixin(PandasDelegate):
         boxed in an index, after being returned from the array
     """
     # raw_methods : dispatch methods that shouldn't be boxed in an Index
-    _raw_methods = set()
+    _raw_methods = set()  # type: Set[str]
     # raw_properties : dispatch properties that shouldn't be boxed in an Index
-    _raw_properties = set()
+    _raw_properties = set()  # type: Set[str]
     name = None
     _data = None
 

pandas/core/internals/blocks.py

@@ -4,10 +4,12 @@
 import inspect
 import re
 import warnings
+from typing import Any, List, Optional
 
 import numpy as np
 
-from pandas._libs import internals as libinternals, lib, tslib, tslibs
+from pandas._libs import (internals as libinternals,  # type: ignore
+                          lib, tslib, tslibs)
 from pandas._libs.tslibs import Timedelta, conversion, is_null_datetimelike
 import pandas.compat as compat
 from pandas.compat import range, zip
@@ -1826,8 +1828,12 @@ def interpolate(self, method='pad', axis=0, inplace=False, limit=None,
                                  limit=limit),
             placement=self.mgr_locs)
 
-    def shift(self, periods, axis=0, fill_value=None):
-        # type: (int, Optional[BlockPlacement], Any) -> List[ExtensionBlock]
+    def shift(self,
+              periods,          # type: int
+              axis=0,           # type: Optional[libinternals.BlockPlacement]
+              fill_value=None   # type: Any
+    ):
+        # type: (...) -> List[ExtensionBlock]
         """
         Shift the block by `periods`.