Updated pyarrow dep to 0.7.0

dhirschfeld · dhirschfeld · commit 44b0fe8231bb · 2017-12-06T22:04:20.000+10:00
Addressed review comments
diff --git a/ci/requirements-2.7.sh b/ci/requirements-2.7.sh
@@ -4,4 +4,4 @@ source activate pandas
 
 echo "install 27"
 
-conda install -n pandas -c conda-forge feather-format pyarrow=0.4.1 fastparquet
+conda install -n pandas -c conda-forge feather-format pyarrow=0.7.0 fastparquet
diff --git a/ci/requirements-3.5.sh b/ci/requirements-3.5.sh
@@ -8,4 +8,4 @@ echo "install 35"
 conda remove -n pandas python-dateutil --force
 pip install python-dateutil
 
-conda install -n pandas -c conda-forge feather-format pyarrow=0.5.0
+conda install -n pandas -c conda-forge feather-format pyarrow=0.7.0
diff --git a/doc/source/install.rst b/doc/source/install.rst
@@ -233,7 +233,7 @@ Optional Dependencies
 * `xarray <http://xarray.pydata.org>`__: pandas like handling for > 2 dims, needed for converting Panels to xarray objects. Version 0.7.0 or higher is recommended.
 * `PyTables <http://www.pytables.org>`__: necessary for HDF5-based storage. Version 3.0.0 or higher required, Version 3.2.1 or higher highly recommended.
 * `Feather Format <https://github.com/wesm/feather>`__: necessary for feather-based storage, version 0.3.1 or higher.
-* `Apache Parquet <https://parquet.apache.org/>`__, either `pyarrow <http://arrow.apache.org/docs/python/>`__ (>= 0.4.1) or `fastparquet <https://fastparquet.readthedocs.io/en/latest/necessary>`__ (>= 0.0.6) for parquet-based storage. The `snappy <https://pypi.python.org/pypi/python-snappy>`__ and `brotli <https://pypi.python.org/pypi/brotlipy>`__ are available for compression support.
+* `Apache Parquet <https://parquet.apache.org/>`__, either `pyarrow <http://arrow.apache.org/docs/python/>`__ (>= 0.7.0) or `fastparquet <https://fastparquet.readthedocs.io/en/latest/necessary>`__ (>= 0.1.0) for parquet-based storage. The `snappy <https://pypi.python.org/pypi/python-snappy>`__ and `brotli <https://pypi.python.org/pypi/brotlipy>`__ are available for compression support.
 * `SQLAlchemy <http://www.sqlalchemy.org>`__: for SQL database support. Version 0.8.1 or higher recommended. Besides SQLAlchemy, you also need a database specific driver. You can find an overview of supported drivers for each SQL dialect in the `SQLAlchemy docs <http://docs.sqlalchemy.org/en/latest/dialects/index.html>`__. Some common drivers are:
 
   * `psycopg2 <http://initd.org/psycopg/>`__: for PostgreSQL
diff --git a/doc/source/whatsnew/v0.22.0.txt b/doc/source/whatsnew/v0.22.0.txt
@@ -163,7 +163,7 @@ Other Enhancements
 - :func:`pandas.read_clipboard` updated to use qtpy, falling back to PyQt5 and then PyQt4, adding compatibility with Python3 and multiple python-qt bindings (:issue:`17722`)
 - Improved wording of ``ValueError`` raised in :func:`read_csv` when the ``usecols`` argument cannot match all columns. (:issue:`17301`)
 - :func:`DataFrame.corrwith` now silently drops non-numeric columns when passed a Series. Before, an exception was raised (:issue:`18570`).
-
+- Enabled the use of non-default indexes in :func:`DataFrame.to_parquet` where the underlying engine supports it (:issue:`18581`)
 
 .. _whatsnew_0220.api_breaking:
 
diff --git a/pandas/io/parquet.py b/pandas/io/parquet.py
@@ -4,6 +4,7 @@
 from distutils.version import LooseVersion
 from pandas import DataFrame, RangeIndex, Int64Index, get_option
 from pandas.compat import range
+from pandas.core.common import AbstractMethodError
 from pandas.io.common import get_filepath_or_buffer
 
 
@@ -39,54 +40,30 @@ class BaseImpl(object):
     api = None  # module
 
     @staticmethod
-    def _validate_index(df):
-        if not isinstance(df.index, Int64Index):
-            msg = (
-                "parquet does not support serializing {} for the index;"
-                "you can .reset_index() to make the index into column(s)"
-            )
-            raise ValueError(msg.format(type(df.index)))
-        if not df.index.equals(RangeIndex(len(df))):
-            raise ValueError(
-                "parquet does not support serializing a non-default index "
-                "for the index; you can .reset_index() to make the index "
-                "into column(s)"
-            )
-        if df.index.name is not None:
-            raise ValueError(
-                "parquet does not serialize index meta-data "
-                "on a default index"
-            )
-
-    @staticmethod
-    def _validate_columns(df):
+    def validate_dataframe(df):
+        if not isinstance(df, DataFrame):
+            raise ValueError("to_parquet only support IO with DataFrames")
         # must have value column names (strings only)
         if df.columns.inferred_type not in {'string', 'unicode'}:
             raise ValueError("parquet must have string column names")
 
-    def validate_dataframe(self, df):
-        if not isinstance(df, DataFrame):
-            raise ValueError("to_parquet only support IO with DataFrames")
-        self._validate_columns(df)
-        self._validate_index(df)
-
     def write(self, df, path, compression, **kwargs):
-        raise NotImplementedError()
+        raise AbstractMethodError()
 
     def read(self, path, columns=None, **kwargs):
-        raise NotImplementedError()
+        raise AbstractMethodError()
 
 
 class PyArrowImpl(BaseImpl):
 
     def __init__(self):
         # since pandas is a dependency of pyarrow
         # we need to import on first use
-
         try:
             import pyarrow
             import pyarrow.parquet
         except ImportError:
+<<<<<<< HEAD
             raise ImportError("pyarrow is required for parquet support\n\n"
                               "you can install via conda\n"
                               "conda install pyarrow -c conda-forge\n"
@@ -104,32 +81,38 @@ def __init__(self):
         self._pyarrow_lt_050 = LooseVersion(pyarrow.__version__) < '0.5.0'
         self._pyarrow_lt_060 = LooseVersion(pyarrow.__version__) < '0.6.0'
         self._pyarrow_lt_070 = LooseVersion(pyarrow.__version__) < '0.7.0'
+=======
+            raise ImportError(
+                "pyarrow is required for parquet support\n\n"
+                "you can install via conda\n"
+                "conda install pyarrow -c conda-forge\n"
+                "\nor via pip\n"
+                "pip install -U pyarrow\n"
+            )
+        if LooseVersion(pyarrow.__version__) < '0.7.0':
+            raise ImportError(
+                "pyarrow >= 0.4.1 is required for parquet support\n\n"
+                "you can install via conda\n"
+                "conda install pyarrow -c conda-forge\n"
+                "\nor via pip\n"
+                "pip install -U pyarrow\n"
+            )
+>>>>>>> Updated pyarrow dep to 0.7.0
         self.api = pyarrow
 
-    def _validate_index(self, df):
-        # pyarrow >= 0.7.0 supports multi-indexes so no need to validate
-        if self._pyarrow_lt_070:
-            super(PyArrowImpl, self)._validate_index(df)
-
     def write(self, df, path, compression='snappy',
               coerce_timestamps='ms', **kwargs):
         self.validate_dataframe(df)
         path, _, _ = get_filepath_or_buffer(path)
-        if self._pyarrow_lt_060:
-            table = self.api.Table.from_pandas(df, timestamps_to_ms=True)
-            self.api.parquet.write_table(
-                table, path, compression=compression, **kwargs)
-
-        else:
-            table = self.api.Table.from_pandas(df)
-            self.api.parquet.write_table(
-                table, path, compression=compression,
-                coerce_timestamps=coerce_timestamps, **kwargs)
+        table = self.api.Table.from_pandas(df)
+        self.api.parquet.write_table(
+            table, path, compression=compression,
+            coerce_timestamps=coerce_timestamps, **kwargs)
 
     def read(self, path, columns=None, **kwargs):
         path, _, _ = get_filepath_or_buffer(path)
-        return self.api.parquet.read_table(path, columns=columns,
-                                           **kwargs).to_pandas()
+        return self.api.parquet.read_table(
+            path, columns=columns, **kwargs).to_pandas()
 
 
 class FastParquetImpl(BaseImpl):
@@ -140,6 +123,7 @@ def __init__(self):
         try:
             import fastparquet
         except ImportError:
+<<<<<<< HEAD
             raise ImportError("fastparquet is required for parquet support\n\n"
                               "you can install via conda\n"
                               "conda install fastparquet -c conda-forge\n"
@@ -154,6 +138,23 @@ def __init__(self):
                               "\nor via pip\n"
                               "pip install -U fastparquet")
 
+=======
+            raise ImportError(
+                "fastparquet is required for parquet support\n\n"
+                "you can install via conda\n"
+                "conda install fastparquet -c conda-forge\n"
+                "\nor via pip\n"
+                "pip install -U fastparquet"
+            )
+        if LooseVersion(fastparquet.__version__) < '0.1.0':
+            raise ImportError(
+                "fastparquet >= 0.1.0 is required for parquet "
+                "support\n\n"
+                "you can install via conda\n"
+                "conda install fastparquet -c conda-forge\n"
+                "\nor via pip\n"
+                "pip install -U fastparquet")
+>>>>>>> Updated pyarrow dep to 0.7.0
         self.api = fastparquet
 
     def write(self, df, path, compression='snappy', **kwargs):
diff --git a/pandas/tests/io/test_parquet.py b/pandas/tests/io/test_parquet.py
@@ -43,6 +43,7 @@ def engine(request):
 def pa():
     if not _HAVE_PYARROW:
         pytest.skip("pyarrow is not installed")
+<<<<<<< HEAD
     return 'pyarrow'
 
 
@@ -60,6 +61,9 @@ def pa_ge_070():
     if not _HAVE_PYARROW:
         pytest.skip("pyarrow is not installed")
     if LooseVersion(pyarrow.__version__) < LooseVersion('0.7.0'):
+=======
+    if LooseVersion(pyarrow.__version__) < '0.7.0':
+>>>>>>> Updated pyarrow dep to 0.7.0
         pytest.skip("pyarrow is < 0.7.0")
     return 'pyarrow'
 
@@ -68,6 +72,8 @@ def pa_ge_070():
 def fp():
     if not _HAVE_FASTPARQUET:
         pytest.skip("fastparquet is not installed")
+    if LooseVersion(fastparquet.__version__) < '0.1.0':
+        pytest.skip("fastparquet is < 0.1.0")
     return 'fastparquet'
 
 
@@ -181,9 +187,7 @@ def test_cross_engine_fp_pa(df_cross_compat, pa, fp):
 class Base(object):
 
     def check_error_on_write(self, df, engine, exc):
-        # check that we are raising the exception
-        # on writing
-
+        # check that we are raising the exception on writing
         with pytest.raises(exc):
             with tm.ensure_clean() as path:
                 to_parquet(df, path, engine, compression=None)
@@ -270,6 +274,32 @@ def test_read_columns(self, engine):
                               write_kwargs={'compression': None},
                               read_kwargs={'columns': ['string']})
 
+    def test_write_with_index(self, engine):
+        df = pd.DataFrame({'A': [1, 2, 3]})
+        self.check_round_trip(df, engine, write_kwargs={'compression': None})
+
+        # non-default index
+        for index in [[2, 3, 4],
+                      pd.date_range('20130101', periods=3),
+                      list('abc'),
+                      [1, 3, 4],
+                      pd.MultiIndex.from_tuples([('a', 1), ('a', 2),
+                                                 ('b', 1)]),
+                      ]:
+
+            df.index = index
+            self.check_round_trip(df, engine)
+
+        # index with meta-data
+        df.index = [0, 1, 2]
+        df.index.name = 'foo'
+        self.check_round_trip(df, engine)
+
+        # column multi-index
+        df.index = [0, 1, 2]
+        df.columns = pd.MultiIndex.from_tuples([('a', 1), ('a', 2), ('b', 1)]),
+        self.check_error_on_write(df, engine, ValueError)
+
 
 class TestParquetPyArrow(Base):
 
@@ -295,14 +325,12 @@ def test_basic(self, pa):
         self.check_round_trip(df, pa)
 
     def test_duplicate_columns(self, pa):
-
         # not currently able to handle duplicate columns
         df = pd.DataFrame(np.arange(12).reshape(4, 3),
                           columns=list('aaa')).copy()
         self.check_error_on_write(df, pa, ValueError)
 
     def test_unsupported(self, pa):
-
         # period
         df = pd.DataFrame({'a': pd.period_range('2013', freq='M', periods=3)})
         self.check_error_on_write(df, pa, ValueError)
@@ -316,78 +344,13 @@ def test_unsupported(self, pa):
         df = pd.DataFrame({'a': ['a', 1, 2.0]})
         self.check_error_on_write(df, pa, ValueError)
 
-    def test_categorical(self, pa_ge_070):
-        pa = pa_ge_070
-
-        # supported in >= 0.7.0
+    def test_categorical(self, pa):
         df = pd.DataFrame({'a': pd.Categorical(list('abc'))})
 
         # de-serialized as object
         expected = df.assign(a=df.a.astype(object))
         self.check_round_trip(df, pa, expected)
 
-    def test_categorical_unsupported(self, pa_lt_070):
-        pa = pa_lt_070
-
-        # supported in >= 0.7.0
-        df = pd.DataFrame({'a': pd.Categorical(list('abc'))})
-        self.check_error_on_write(df, pa, NotImplementedError)
-
-    def test_write_with_index(self, pa_lt_070):
-        engine = pa_lt_070
-        df = pd.DataFrame({'A': [1, 2, 3]})
-        self.check_round_trip(df, engine, write_kwargs={'compression': None})
-
-        # non-default index
-        for index in [[2, 3, 4],
-                      pd.date_range('20130101', periods=3),
-                      list('abc'),
-                      [1, 3, 4],
-                      pd.MultiIndex.from_tuples([('a', 1), ('a', 2),
-                                                 ('b', 1)]),
-                      ]:
-
-            df.index = index
-            self.check_error_on_write(df, engine, ValueError)
-
-        # index with meta-data
-        df.index = [0, 1, 2]
-        df.index.name = 'foo'
-        self.check_error_on_write(df, engine, ValueError)
-
-        # column multi-index
-        df.index = [0, 1, 2]
-        df.columns = pd.MultiIndex.from_tuples([('a', 1), ('a', 2), ('b', 1)]),
-        self.check_error_on_write(df, engine, ValueError)
-
-    def test_write_with_non_default_index(self, pa_ge_070):
-        engine = pa_ge_070
-
-        df = pd.DataFrame({'A': [1, 2, 3]})
-        self.check_round_trip(df, engine, write_kwargs={'compression': None})
-
-        # non-default index
-        for index in [[2, 3, 4],
-                      pd.date_range('20130101', periods=3),
-                      list('abc'),
-                      [1, 3, 4],
-                      pd.MultiIndex.from_tuples([('a', 1), ('a', 2),
-                                                 ('b', 1)]),
-                      ]:
-
-            df.index = index
-            self.check_round_trip(df, engine)
-
-        # index with meta-data
-        df.index = [0, 1, 2]
-        df.index.name = 'foo'
-        self.check_round_trip(df, engine)
-
-        # column multi-index
-        df.index = [0, 1, 2]
-        df.columns = pd.MultiIndex.from_tuples([('a', 1), ('a', 2), ('b', 1)]),
-        self.check_error_on_write(df, engine, ValueError)
-
 
 class TestParquetFastParquet(Base):
 
@@ -413,34 +376,6 @@ def test_basic(self, fp):
 
         self.check_round_trip(df, fp, write_kwargs={'compression': None})
 
-    def test_write_with_index(self, fp):
-        engine = fp
-
-        df = pd.DataFrame({'A': [1, 2, 3]})
-        self.check_round_trip(df, engine, write_kwargs={'compression': None})
-
-        # non-default index
-        for index in [[2, 3, 4],
-                      pd.date_range('20130101', periods=3),
-                      list('abc'),
-                      [1, 3, 4],
-                      pd.MultiIndex.from_tuples([('a', 1), ('a', 2),
-                                                 ('b', 1)]),
-                      ]:
-
-            df.index = index
-            self.check_error_on_write(df, engine, ValueError)
-
-        # index with meta-data
-        df.index = [0, 1, 2]
-        df.index.name = 'foo'
-        self.check_error_on_write(df, engine, ValueError)
-
-        # column multi-index
-        df.index = [0, 1, 2]
-        df.columns = pd.MultiIndex.from_tuples([('a', 1), ('a', 2), ('b', 1)]),
-        self.check_error_on_write(df, engine, ValueError)
-
     @pytest.mark.skip(reason="not supported")
     def test_duplicate_columns(self, fp):
 

Original file line number	Diff line number	Diff line change
`@@ -4,4 +4,4 @@ source activate pandas`
`4`	`4`
`5`	`5`	`echo "install 27"`
`6`	`6`
`7`		`-conda install -n pandas -c conda-forge feather-format pyarrow=0.4.1 fastparquet`
	`7`	`+conda install -n pandas -c conda-forge feather-format pyarrow=0.7.0 fastparquet`