How to use pandas_version method in pandera

Best Python code snippet using pandera_python

test_json_table_schema_ext_dtype.py

Source:test_json_table_schema_ext_dtype.py

1"""Tests for ExtensionDtype Table Schema integration."""2from collections import OrderedDict3import datetime as dt4import decimal5import json6import pytest7from pandas import (8    DataFrame,9    array,10)11from pandas.core.arrays.integer import Int64Dtype12from pandas.core.arrays.string_ import StringDtype13from pandas.core.series import Series14from pandas.tests.extension.date import (15    DateArray,16    DateDtype,17)18from pandas.tests.extension.decimal.array import (19    DecimalArray,20    DecimalDtype,21)22from pandas.io.json._table_schema import (23    as_json_table_type,24    build_table_schema,25)26class TestBuildSchema:27    def setup_method(self, method):28        self.da = DateArray([dt.date(2021, 10, 10)])29        self.dc = DecimalArray([decimal.Decimal(10)])30        self.sa = array(["pandas"], dtype="string")31        self.ia = array([10], dtype="Int64")32        self.df = DataFrame(33            {34                "A": self.da,35                "B": self.dc,36                "C": self.sa,37                "D": self.ia,38            }39        )40    def test_build_table_schema(self):41        result = build_table_schema(self.df, version=False)42        expected = {43            "fields": [44                {"name": "index", "type": "integer"},45                {"name": "A", "type": "any", "extDtype": "DateDtype"},46                {"name": "B", "type": "any", "extDtype": "decimal"},47                {"name": "C", "type": "any", "extDtype": "string"},48                {"name": "D", "type": "integer", "extDtype": "Int64"},49            ],50            "primaryKey": ["index"],51        }52        assert result == expected53        result = build_table_schema(self.df)54        assert "pandas_version" in result55class TestTableSchemaType:56    @pytest.mark.parametrize(57        "date_data",58        [59            DateArray([dt.date(2021, 10, 10)]),60            DateArray(dt.date(2021, 10, 10)),61            Series(DateArray(dt.date(2021, 10, 10))),62        ],63    )64    def test_as_json_table_type_ext_date_array_dtype(self, date_data):65        assert as_json_table_type(date_data.dtype) == "any"66    def test_as_json_table_type_ext_date_dtype(self):67        assert as_json_table_type(DateDtype()) == "any"68    @pytest.mark.parametrize(69        "decimal_data",70        [71            DecimalArray([decimal.Decimal(10)]),72            Series(DecimalArray([decimal.Decimal(10)])),73        ],74    )75    def test_as_json_table_type_ext_decimal_array_dtype(self, decimal_data):76        assert as_json_table_type(decimal_data.dtype) == "any"77    def test_as_json_table_type_ext_decimal_dtype(self):78        assert as_json_table_type(DecimalDtype()) == "any"79    @pytest.mark.parametrize(80        "string_data",81        [82            array(["pandas"], dtype="string"),83            Series(array(["pandas"], dtype="string")),84        ],85    )86    def test_as_json_table_type_ext_string_array_dtype(self, string_data):87        assert as_json_table_type(string_data.dtype) == "any"88    def test_as_json_table_type_ext_string_dtype(self):89        assert as_json_table_type(StringDtype()) == "any"90    @pytest.mark.parametrize(91        "integer_data",92        [93            array([10], dtype="Int64"),94            Series(array([10], dtype="Int64")),95        ],96    )97    def test_as_json_table_type_ext_integer_array_dtype(self, integer_data):98        assert as_json_table_type(integer_data.dtype) == "integer"99    def test_as_json_table_type_ext_integer_dtype(self):100        assert as_json_table_type(Int64Dtype()) == "integer"101class TestTableOrient:102    def setup_method(self, method):103        self.da = DateArray([dt.date(2021, 10, 10)])104        self.dc = DecimalArray([decimal.Decimal(10)])105        self.sa = array(["pandas"], dtype="string")106        self.ia = array([10], dtype="Int64")107        self.df = DataFrame(108            {109                "A": self.da,110                "B": self.dc,111                "C": self.sa,112                "D": self.ia,113            }114        )115    def test_build_date_series(self):116        s = Series(self.da, name="a")117        s.index.name = "id"118        result = s.to_json(orient="table", date_format="iso")119        result = json.loads(result, object_pairs_hook=OrderedDict)120        assert "pandas_version" in result["schema"]121        result["schema"].pop("pandas_version")122        fields = [123            {"name": "id", "type": "integer"},124            {"name": "a", "type": "any", "extDtype": "DateDtype"},125        ]126        schema = {"fields": fields, "primaryKey": ["id"]}127        expected = OrderedDict(128            [129                ("schema", schema),130                ("data", [OrderedDict([("id", 0), ("a", "2021-10-10T00:00:00.000Z")])]),131            ]132        )133        assert result == expected134    def test_build_decimal_series(self):135        s = Series(self.dc, name="a")136        s.index.name = "id"137        result = s.to_json(orient="table", date_format="iso")138        result = json.loads(result, object_pairs_hook=OrderedDict)139        assert "pandas_version" in result["schema"]140        result["schema"].pop("pandas_version")141        fields = [142            {"name": "id", "type": "integer"},143            {"name": "a", "type": "any", "extDtype": "decimal"},144        ]145        schema = {"fields": fields, "primaryKey": ["id"]}146        expected = OrderedDict(147            [148                ("schema", schema),149                ("data", [OrderedDict([("id", 0), ("a", 10.0)])]),150            ]151        )152        assert result == expected153    def test_build_string_series(self):154        s = Series(self.sa, name="a")155        s.index.name = "id"156        result = s.to_json(orient="table", date_format="iso")157        result = json.loads(result, object_pairs_hook=OrderedDict)158        assert "pandas_version" in result["schema"]159        result["schema"].pop("pandas_version")160        fields = [161            {"name": "id", "type": "integer"},162            {"name": "a", "type": "any", "extDtype": "string"},163        ]164        schema = {"fields": fields, "primaryKey": ["id"]}165        expected = OrderedDict(166            [167                ("schema", schema),168                ("data", [OrderedDict([("id", 0), ("a", "pandas")])]),169            ]170        )171        assert result == expected172    def test_build_int64_series(self):173        s = Series(self.ia, name="a")174        s.index.name = "id"175        result = s.to_json(orient="table", date_format="iso")176        result = json.loads(result, object_pairs_hook=OrderedDict)177        assert "pandas_version" in result["schema"]178        result["schema"].pop("pandas_version")179        fields = [180            {"name": "id", "type": "integer"},181            {"name": "a", "type": "integer", "extDtype": "Int64"},182        ]183        schema = {"fields": fields, "primaryKey": ["id"]}184        expected = OrderedDict(185            [186                ("schema", schema),187                ("data", [OrderedDict([("id", 0), ("a", 10)])]),188            ]189        )190        assert result == expected191    def test_to_json(self):192        df = self.df.copy()193        df.index.name = "idx"194        result = df.to_json(orient="table", date_format="iso")195        result = json.loads(result, object_pairs_hook=OrderedDict)196        assert "pandas_version" in result["schema"]197        result["schema"].pop("pandas_version")198        fields = [199            OrderedDict({"name": "idx", "type": "integer"}),200            OrderedDict({"name": "A", "type": "any", "extDtype": "DateDtype"}),201            OrderedDict({"name": "B", "type": "any", "extDtype": "decimal"}),202            OrderedDict({"name": "C", "type": "any", "extDtype": "string"}),203            OrderedDict({"name": "D", "type": "integer", "extDtype": "Int64"}),204        ]205        schema = OrderedDict({"fields": fields, "primaryKey": ["idx"]})206        data = [207            OrderedDict(208                [209                    ("idx", 0),210                    ("A", "2021-10-10T00:00:00.000Z"),211                    ("B", 10.0),212                    ("C", "pandas"),213                    ("D", 10),214                ]215            )216        ]217        expected = OrderedDict([("schema", schema), ("data", data)])...

_compat.py

Source:_compat.py

1import string2import numpy as np3import pandas as pd4from packaging.version import parse as parse_version5PANDAS_VERSION = parse_version(pd.__version__)6PANDAS_GT_104 = PANDAS_VERSION >= parse_version("1.0.4")7PANDAS_GT_110 = PANDAS_VERSION >= parse_version("1.1.0")8PANDAS_GT_120 = PANDAS_VERSION >= parse_version("1.2.0")9PANDAS_GT_121 = PANDAS_VERSION >= parse_version("1.2.1")10PANDAS_GT_130 = PANDAS_VERSION >= parse_version("1.3.0")11PANDAS_GT_131 = PANDAS_VERSION >= parse_version("1.3.1")12PANDAS_GT_133 = PANDAS_VERSION >= parse_version("1.3.3")13PANDAS_GT_140 = PANDAS_VERSION >= parse_version("1.4.0")14# FIXME: Using `.release` below as versions like `1.5.0.dev0+268.gbe8d1ec880`15# are less than `1.5.0` with `packaging.version`. Update to use `parse_version("1.5.0")`16# below once `pandas=1.5.0` is released17PANDAS_GT_150 = PANDAS_VERSION.release >= (1, 5, 0)18import pandas.testing as tm19def assert_categorical_equal(left, right, *args, **kwargs):20    tm.assert_extension_array_equal(left, right, *args, **kwargs)21    assert pd.api.types.is_categorical_dtype(22        left.dtype23    ), f"{left} is not categorical dtype"24    assert pd.api.types.is_categorical_dtype(25        right.dtype26    ), f"{right} is not categorical dtype"27def assert_numpy_array_equal(left, right):28    left_na = pd.isna(left)29    right_na = pd.isna(right)30    np.testing.assert_array_equal(left_na, right_na)31    left_valid = left[~left_na]32    right_valid = right[~right_na]33    np.testing.assert_array_equal(left_valid, right_valid)34def makeDataFrame():35    data = np.random.randn(30, 4)36    index = list(string.ascii_letters)[:30]37    return pd.DataFrame(data, index=index, columns=list("ABCD"))38def makeTimeDataFrame():39    data = makeDataFrame()40    data.index = makeDateIndex()41    return data42def makeTimeSeries():43    return makeTimeDataFrame()["A"]44def makeDateIndex(k=30, freq="B"):45    return pd.date_range("2000", periods=k, freq=freq)46def makeTimedeltaIndex(k=30, freq="D"):47    return pd.timedelta_range("1 day", periods=k, freq=freq)48def makeMissingDataframe():49    df = makeDataFrame()50    data = df.values51    data = np.where(data > 1, np.nan, data)52    return pd.DataFrame(data, index=df.index, columns=df.columns)53def makeMixedDataFrame():54    df = pd.DataFrame(55        {56            "A": [0.0, 1, 2, 3, 4],57            "B": [0.0, 1, 0, 1, 0],58            "C": [f"foo{i}" for i in range(5)],59            "D": pd.date_range("2009-01-01", periods=5),60        }61    )...

_utils.py

Source:_utils.py

1import datetime as dt2from distutils.version import LooseVersion3import pandas as pd4from pandas.core.common import PandasError5from pandas import to_datetime6import requests7from requests_file import FileAdapter8import requests_ftp9requests_ftp.monkeypatch_session()10if pd.compat.PY3:11    from urllib.error import HTTPError     # noqa12else:13    from urllib2 import HTTPError          # noqa14PANDAS_VERSION = LooseVersion(pd.__version__)15if PANDAS_VERSION >= LooseVersion('0.19.0'):16    PANDAS_0190 = True17    from pandas.api.types import is_number                   # noqa18else:19    PANDAS_0190 = False20    from pandas.core.common import is_number                 # noqa21if PANDAS_VERSION >= LooseVersion('0.17.0'):22    PANDAS_0170 = True23else:24    PANDAS_0170 = False25if PANDAS_VERSION >= LooseVersion('0.16.0'):26    PANDAS_0160 = True27else:28    PANDAS_0160 = False29if PANDAS_VERSION >= LooseVersion('0.14.0'):30    PANDAS_0140 = True31else:32    PANDAS_0140 = False33class SymbolWarning(UserWarning):34    pass35class RemoteDataError(PandasError, IOError):36    pass37def _sanitize_dates(start, end):38    """39    Return (datetime_start, datetime_end) tuple40    if start is None - default is 2010/01/0141    if end is None - default is today42    """43    if is_number(start):44        # regard int as year45        start = dt.datetime(start, 1, 1)46    start = to_datetime(start)47    if is_number(end):48        end = dt.datetime(end, 1, 1)49    end = to_datetime(end)50    if start is None:51        start = dt.datetime(2010, 1, 1)52    if end is None:53        end = dt.datetime.today()54    return start, end55def _init_session(session, retry_count=3):56    if session is None:57        session = requests.Session()58        session.mount('file://', FileAdapter())59        # do not set requests max_retries here to support arbitrary pause...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.