Add setitem (#394)

* Impl operator setitem

* change

* Imp operator setitem for Series

* comment old style

* change

* change

* Add conditions in pd_series_ext for operator setitem

* change

* change

* change

* fix code style issues

* change

* change and add example for setitem (now not work)

* change

* change

* change

* change

* change

* change

* change

* change comment in examples

* deleted brackets in message error for operator setitem

* error message extension

* solution PEP 8 issues

* correctioned error message in test

* add template for message error

* correction examples, expansion tests

* correction tests

* correction check

Author: Rubtsowa

examples/series_setitem_int.py

@@ -0,0 +1,43 @@
+# *****************************************************************************
+# Copyright (c) 2019, Intel Corporation All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+#
+#     Redistributions of source code must retain the above copyright notice,
+#     this list of conditions and the following disclaimer.
+#
+#     Redistributions in binary form must reproduce the above copyright notice,
+#     this list of conditions and the following disclaimer in the documentation
+#     and/or other materials provided with the distribution.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
+# THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
+# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
+# OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+# WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+# OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
+# EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+# *****************************************************************************
+
+import numpy as np
+import pandas as pd
+
+from numba import njit
+
+
+@njit
+def series_setitem():
+    value = 0
+    series = pd.Series(np.arange(5, 0, -1))  # Series of 5, 4, 3, 2, 1
+
+    series[0] = value
+
+    return series   # result Series of 0, 4, 3, 2, 1
+
+
+print(series_setitem())

examples/series_setitem_series.py

@@ -0,0 +1,44 @@
+# *****************************************************************************
+# Copyright (c) 2019, Intel Corporation All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+#
+#     Redistributions of source code must retain the above copyright notice,
+#     this list of conditions and the following disclaimer.
+#
+#     Redistributions in binary form must reproduce the above copyright notice,
+#     this list of conditions and the following disclaimer in the documentation
+#     and/or other materials provided with the distribution.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
+# THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
+# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
+# OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+# WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+# OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
+# EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+# *****************************************************************************
+
+import numpy as np
+import pandas as pd
+
+from numba import njit
+
+
+@njit
+def series_setitem():
+    value = 0
+    series = pd.Series(np.arange(5, 0, -1))  # Series of 5, 4, 3, 2, 1
+
+    indices = pd.Series(np.asarray([1, 3]))
+    series[indices] = value
+
+    return series       # result Series of 5, 0, 3, 0, 1
+
+
+print(series_setitem())

examples/series_setitem_slice.py

@@ -0,0 +1,43 @@
+# *****************************************************************************
+# Copyright (c) 2019, Intel Corporation All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+#
+#     Redistributions of source code must retain the above copyright notice,
+#     this list of conditions and the following disclaimer.
+#
+#     Redistributions in binary form must reproduce the above copyright notice,
+#     this list of conditions and the following disclaimer in the documentation
+#     and/or other materials provided with the distribution.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
+# THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
+# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
+# OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+# WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+# OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
+# EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+# *****************************************************************************
+
+import numpy as np
+import pandas as pd
+
+from numba import njit
+
+
+@njit
+def series_setitem():
+    value = 0
+    series = pd.Series(np.arange(5, 0, -1))  # Series of 5, 4, 3, 2, 1
+
+    series[2:5] = value
+
+    return series   # result Series of 5, 4, 0, 0, 0
+
+
+print(series_setitem())

sdc/datatypes/hpat_pandas_series_functions.py

@@ -141,6 +141,104 @@ def hpat_pandas_series_getitem_idx_series_impl(self, idx):
     raise TypingError('{} The index must be an Integer, Slice or a pandas.series. Given: {}'.format(_func_name, idx))
 
 
+@sdc_overload(operator.setitem)
+def hpat_pandas_series_setitem(self, idx, value):
+    """
+    Intel Scalable Dataframe Compiler User Guide
+    ********************************************
+    Pandas API: pandas.Series.set
+
+    Examples
+    --------
+    .. literalinclude:: ../../../examples/series_setitem_int.py
+       :language: python
+       :lines: 27-
+       :caption: Setting Pandas Series elements
+       :name: ex_series_setitem
+
+    .. code-block:: console
+
+        > python ./series_setitem_int.py
+
+            0    0
+            1    4
+            2    3
+            3    2
+            4    1
+            dtype: int64
+
+        > python ./series_setitem_slice.py
+
+            0    5
+            1    4
+            2    0
+            3    0
+            4    0
+            dtype: int64
+
+        > python ./series_setitem_series.py
+
+            0    5
+            1    0
+            2    3
+            3    0
+            4    1
+            dtype: int64
+
+    Intel Scalable Dataframe Compiler Developer Guide
+    *************************************************
+     Pandas Series operator :attr:`pandas.Series.set` implementation
+
+    Test: python -m sdc.runtests -k sdc.tests.test_series.TestSeries.test_series_setitem*
+
+    Parameters
+    ----------
+    series: :obj:`pandas.Series`
+        input series
+    idx: :obj:`int`, :obj:`slice` or :obj:`pandas.Series`
+        input index
+    value: :object
+        input value
+
+    Returns
+    -------
+    :class:`pandas.Series` or an element of the underneath type
+            object of :class:`pandas.Series`
+    """
+
+    ty_checker = TypeChecker('Operator setitem.')
+    ty_checker.check(self, SeriesType)
+
+    if not (isinstance(idx, (types.Integer, types.SliceType, SeriesType))):
+        ty_checker.raise_exc(idx, 'int, Slice, Series', 'idx')
+
+    if not((isinstance(value, SeriesType) and isinstance(value.dtype, self.dtype)) or \
+           isinstance(value, type(self.dtype))):
+        ty_checker.raise_exc(value, self.dtype, 'value')
+
+    if isinstance(idx, types.Integer) or isinstance(idx, types.SliceType):
+        def hpat_pandas_series_setitem_idx_integer_impl(self, idx, value):
+            """
+            Test: python -m sdc.runtests sdc.tests.test_series.TestSeries.test_series_setitem_for_value
+            Test: python -m sdc.runtests sdc.tests.test_series.TestSeries.test_series_setitem_for_slice
+            """
+            self._data[idx] = value
+            return self
+
+        return hpat_pandas_series_setitem_idx_integer_impl
+
+    if isinstance(idx, SeriesType):
+        def hpat_pandas_series_setitem_idx_series_impl(self, idx, value):
+            """
+            Test: python -m sdc.runtests sdc.tests.test_series.TestSeries.test_series_setitem_for_series
+            """
+            super_index = idx._data
+            self._data[super_index] = value
+            return self
+
+        return hpat_pandas_series_setitem_idx_series_impl
+
+
 @sdc_overload_attribute(SeriesType, 'at')
 @sdc_overload_attribute(SeriesType, 'iat')
 @sdc_overload_attribute(SeriesType, 'iloc')

sdc/hiframes/pd_series_ext.py

@@ -965,21 +965,22 @@ class CmpOpLESeries(SeriesCompEqual):
 class CmpOpLTSeries(SeriesCompEqual):
     key = '<'
 
-# @infer_global(operator.getitem)
-# class GetItemBuffer(AbstractTemplate):
-#     key = operator.getitem
 
-#     def generic(self, args, kws):
-#         assert not kws
-#         [ary, idx] = args
-#         import pdb; pdb.set_trace()
-#         if not isinstance(ary, SeriesType):
-#             return
-#         out = get_array_index_type(ary, idx)
-#         # check result to be dt64 since it might be sliced array
-#         # replace result with Timestamp
-#         if out is not None and out.result == types.NPDatetime('ns'):
-#             return signature(pandas_timestamp_type, ary, out.index)
+if sdc.config.config_pipeline_hpat_default:
+    @infer_global(operator.getitem)
+    class GetItemBuffer(AbstractTemplate):
+        key = operator.getitem
+
+        def generic(self, args, kws):
+            assert not kws
+            [ary, idx] = args
+            if not isinstance(ary, SeriesType):
+                return
+            out = get_array_index_type(ary, idx)
+            # check result to be dt64 since it might be sliced array
+            # replace result with Timestamp
+            if out is not None and out.result == types.NPDatetime('ns'):
+                return signature(pandas_timestamp_type, ary, out.index)
 
 
 def install_array_method(name, generic):
@@ -1105,44 +1106,44 @@ def generic(self, args, kws):
                     sig.return_type = pandas_timestamp_type
             return sig
 
+if sdc.config.config_pipeline_hpat_default:
+    @infer_global(operator.setitem)
+    class SetItemSeries(SetItemBuffer):
+        def generic(self, args, kws):
+            assert not kws
+            series, idx, val = args
+            if not isinstance(series, SeriesType):
+                return None
+            # TODO: handle any of args being Series independently
+            ary = series_to_array_type(series)
+            is_idx_series = False
+            if isinstance(idx, SeriesType):
+                idx = series_to_array_type(idx)
+                is_idx_series = True
+            is_val_series = False
+            if isinstance(val, SeriesType):
+                val = series_to_array_type(val)
+                is_val_series = True
+            # TODO: strings, dt_index
+            res = super(SetItemSeries, self).generic((ary, idx, val), kws)
+            if res is not None:
+                new_series = if_arr_to_series_type(res.args[0])
+                idx = res.args[1]
+                val = res.args[2]
+                if is_idx_series:
+                    idx = if_arr_to_series_type(idx)
+                if is_val_series:
+                    val = if_arr_to_series_type(val)
+                res.args = (new_series, idx, val)
+                return res
 
-@infer_global(operator.setitem)
-class SetItemSeries(SetItemBuffer):
-    def generic(self, args, kws):
-        assert not kws
-        series, idx, val = args
-        if not isinstance(series, SeriesType):
-            return None
-        # TODO: handle any of args being Series independently
-        ary = series_to_array_type(series)
-        is_idx_series = False
-        if isinstance(idx, SeriesType):
-            idx = series_to_array_type(idx)
-            is_idx_series = True
-        is_val_series = False
-        if isinstance(val, SeriesType):
-            val = series_to_array_type(val)
-            is_val_series = True
-        # TODO: strings, dt_index
-        res = super(SetItemSeries, self).generic((ary, idx, val), kws)
-        if res is not None:
-            new_series = if_arr_to_series_type(res.args[0])
-            idx = res.args[1]
-            val = res.args[2]
-            if is_idx_series:
-                idx = if_arr_to_series_type(idx)
-            if is_val_series:
-                val = if_arr_to_series_type(val)
-            res.args = (new_series, idx, val)
-            return res
-
-
-@infer_global(operator.setitem)
-class SetItemSeriesIat(SetItemSeries):
-    def generic(self, args, kws):
-        # iat[] is the same as regular setitem
-        if isinstance(args[0], SeriesIatType):
-            return SetItemSeries.generic(self, (args[0].stype, args[1], args[2]), kws)
+if sdc.config.config_pipeline_hpat_default:
+    @infer_global(operator.setitem)
+    class SetItemSeriesIat(SetItemSeries):
+        def generic(self, args, kws):
+            # iat[] is the same as regular setitem
+            if isinstance(args[0], SeriesIatType):
+                return SetItemSeries.generic(self, (args[0].stype, args[1], args[2]), kws)
 
 
 inplace_ops = [