PEP8 whitespace fix

Author: detrout

pandas/tests/io/test_ods.py

@@ -0,0 +1,149 @@
+from collections import OrderedDict
+import os
+from numpy import nan
+import pandas
+import pandas.util._test_decorators as td
+import pandas.util.testing as tm
+from pandas import DataFrame, Timestamp, Timedelta
+from pandas.io.opendocument import ODFReader
+import pytest
+
+
+@td.skip_if_no('odf')
+class TestOpenDocument(object):
+    @pytest.fixture(autouse=True)
+    def setup_method(self, datapath):
+        self.dirpath = datapath("io", "data")
+
+    def get_opendocument(self, filename, *args, **kwargs):
+        """
+        Return ODFReader class containing tables from parsed OpenDocument file
+
+        Parameters
+        ----------
+        filename : str
+
+        Returns
+        -------
+
+        document : ODFReader object
+        """
+        pth = os.path.join(self.dirpath, filename)
+        document = ODFReader(pth)
+        return document
+
+    def get_opendocumentdf(self, filename, *args, **kwargs):
+        """
+        Return DataFrame from named sheet in a parsed OpenDocument file
+
+        Parameters
+        ----------
+        filename : str
+            File base name
+
+        Returns
+        -------
+
+        df : DataFrame
+        """
+        document = self.get_opendocument(filename)
+        return document.parse(*args, **kwargs)
+
+    def test_read_types(self):
+        """Make sure we read ODF data types correctly
+        """
+        book = self.get_opendocument('datatypes.ods')
+        assert len(book.sheet_names) == 1
+        assert book.sheet_names == ['Sheet1']
+        sheet = book.parse('Sheet1', header=None)
+
+        expected = DataFrame(
+            [[1.0],
+             [1.25],
+             ['a'],
+             [Timestamp(2003, 1, 2)],
+             [False],
+             [0.35],
+             [Timedelta(hours=3, minutes=45),
+              Timedelta(hours=17, minutes=53),
+              Timedelta(hours=14, minutes=8)],
+             # though what should the value of a hyperlink be?
+             ['UBERON:0002101']])
+        tm.assert_equal(sheet, expected)
+
+    def test_read_lower_diagonal(self):
+        """TextParser failed when given an irregular list of lists
+
+        Make sure we can parse:
+        1
+        2 3
+        4 5 6
+        7 8 9 10
+        """
+        sheet = self.get_opendocumentdf(
+            'lowerdiagonal.ods', 'Sheet1',
+            index_col=None, header=None)
+
+        assert sheet.shape == (4, 4)
+
+    def test_read_headers(self):
+        """Do we read headers correctly?
+        """
+        sheet = self.get_opendocumentdf(
+            'headers.ods', 'Sheet1', index_col=0)
+
+        expected = DataFrame.from_dict(OrderedDict([
+            ("Header", ["Row 1", "Row 2"]),
+            ("Column 1", [1.0, 2.0]),
+            ("Column 2", [3.0, 4.0]),
+            # Empty Column
+            ("Column 4", [7.0, 8.0]),
+            # Empty Column 2
+            ("Column 6", [11.0, 12.0])]))
+        expected.set_index("Header", inplace=True)
+        columns = ["Column 1", "Column 2", "Column 4", "Column 6"]
+        tm.assert_equal(sheet[columns], expected)
+        empties = [None, 'None.1']
+        for name in empties:
+            for value in sheet[name]:
+                assert pandas.isnull(value)
+
+    def test_read_writer_table(self):
+        """ODF reuses the same table tags in Writer and Presentation files
+
+        Test reading a table out of a text document
+        """
+        table = self.get_opendocumentdf(
+            'writertable.odt', 'Table1', index_col=0)
+
+        assert table.shape == (3, 3)
+        expected = DataFrame.from_dict(OrderedDict([
+            ("Header", ["Row 1", "Row 2", "Row 3"]),
+            ("Column 1", [1.0, 2.0, 3.0]),
+            ("Unnamed: 2", [nan, nan, nan]),
+            ("Column 3", [7.0, 8.0, 9.0])]))
+        expected.set_index("Header", inplace=True)
+        columns = ["Column 1", "Column 3"]
+        tm.assert_equal(table[columns], expected[columns])
+
+        # make sure pandas gives a name to the unnamed column
+        for i in range(3):
+            assert pandas.isnull(table["Unnamed: 2"][i])
+
+    def test_blank_row_repeat(self):
+        table = self.get_opendocumentdf(
+            'blank-row-repeat.ods', 'Biosamples')
+
+        assert table.shape == (8, 9)
+        assert table['biosample_accession'][7] == 9.0
+
+    def test_runlengthencoding(self):
+        """Calc will use repeat when adjacent columns have the same value.
+        """
+        sheet = self.get_opendocumentdf(
+            'runlengthencoding.ods', 'Sheet1', header=None)
+        assert sheet.shape == (5, 3)
+        # check by column, not by row.
+        assert list(sheet[0]) == [1.0, 1.0, 2.0, 2.0, 2.0]
+        assert list(sheet[1]) == [1.0, 2.0, 2.0, 2.0, 2.0]
+        assert list(sheet[2]) == [1.0, 2.0, 2.0, 2.0, 2.0]