def test_dataframe_multiindex():
"""Test to validate proper creation and handling of a multiindex."""
arrays = [numpy.array(['bar', 'bar', 'baz', 'baz',
'foo', 'foo', 'qux', 'qux']),
numpy.array(['one', 'two', 'one', 'two',
'one', 'two', 'one', 'two'])]
tuples = list(zip(*arrays))
index = MultiIndex.from_tuples(tuples, names=['first', 'second'])
df = DataFrame(numpy.random.randn(6, 6), index=index[:6],
columns=index[:6])
editor = DataFrameEditor(None)
editor.setup_and_check(df)
header = editor.table_header.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == 0
assert data_header(header, 0, 0) == 'bar'
assert data_header(header, 1, 0) == 'one'
assert data_header(header, 0, 1) == 'bar'
assert data_header(header, 1, 1) == 'two'
assert data_header(header, 0, 2) == 'baz'
assert data_header(header, 1, 2) == 'one'
assert data_header(header, 0, 3) == 'baz'
assert data_header(header, 1, 3) == 'two'
assert data_header(header, 0, 4) == 'foo'
assert data_header(header, 1, 4) == 'one'
assert data_header(header, 0, 5) == 'foo'
assert data_header(header, 1, 5) == 'two'
def test_header_bom():
"""Test for BOM data in the headers."""
df = read_csv(os.path.join(FILES_PATH, 'issue_2514.csv'))
editor = DataFrameEditor(None)
editor.setup_and_check(df)
header = editor.table_header.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == "Date (MMM-YY)"
def test_change_format_emits_signal(qtbot, monkeypatch):
mockQInputDialog = Mock()
mockQInputDialog.getText = lambda parent, title, label, mode, text: ('%10.3e', True)
monkeypatch.setattr('spyder.widgets.variableexplorer.dataframeeditor.QInputDialog', mockQInputDialog)
df = DataFrame([[0]])
editor = DataFrameEditor(None)
editor.setup_and_check(df)
with qtbot.waitSignal(editor.sig_option_changed) as blocker:
editor.change_format()
assert blocker.args == ['dataframe_format', '%10.3e']
def test_dataframe_simpleindex(qtbot):
"""Test to validate proper creation and handling of a simpleindex."""
df = DataFrame(numpy.random.randn(6, 6))
editor = DataFrameEditor(None)
editor.setup_and_check(df)
header = editor.table_header.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == "0"
assert header.headerData(1, Qt.Horizontal,
Qt.DisplayRole) == "1"
assert header.headerData(5, Qt.Horizontal,
Qt.DisplayRole) == "5"
def test_sort_dataframe_with_category_dtypes(qtbot): # cf. issue 5361
df = DataFrame({'A': [1, 2, 3, 4],
'B': ['a', 'b', 'c', 'd']})
df = df.astype(dtype={'B': 'category'})
df_cols = df.dtypes
editor = DataFrameEditor(None)
editor.setup_and_check(df_cols)
dfm = editor.dataModel
QTimer.singleShot(1000, lambda: close_message_box(qtbot))
editor.dataModel.sort(0)
assert data(dfm, 0, 0) == 'int64'
assert data(dfm, 1, 0) == 'category'
def test_change_format_with_format_not_starting_with_percent(qtbot, monkeypatch):
mockQInputDialog = Mock()
mockQInputDialog.getText = lambda parent, title, label, mode, text: ('xxx%f', True)
monkeypatch.setattr('spyder.widgets.variableexplorer.dataframeeditor'
'.QInputDialog', mockQInputDialog)
monkeypatch.setattr('spyder.widgets.variableexplorer.dataframeeditor'
'.QMessageBox.critical', Mock())
df = DataFrame([[0]])
editor = DataFrameEditor(None)
editor.setup_and_check(df)
with qtbot.assertNotEmitted(editor.sig_option_changed):
editor.change_format()
def test_sort_dataframe_with_duplicate_column(qtbot):
df = DataFrame({'A': [1, 3, 2], 'B': [4, 6, 5]})
df = concat((df, df.A), axis=1)
editor = DataFrameEditor(None)
editor.setup_and_check(df)
dfm = editor.dataModel
QTimer.singleShot(1000, lambda: close_message_box(qtbot))
editor.dataModel.sort(0)
assert [data(dfm, row, 0) for row in range(len(df))] == ['1', '3', '2']
assert [data(dfm, row, 1) for row in range(len(df))] == ['4', '6', '5']
editor.dataModel.sort(1)
assert [data(dfm, row, 0) for row in range(len(df))] == ['1', '2', '3']
assert [data(dfm, row, 1) for row in range(len(df))] == ['4', '5', '6']
def test_dataframe_simpleindex_custom_columns():
"""Test to validate proper creation and handling of custom simpleindex."""
df = DataFrame(numpy.random.randn(10, 5),
columns=['a', 'b', 'c', 'd', 'e'])
editor = DataFrameEditor(None)
editor.setup_and_check(df)
header = editor.table_header.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == "a"
assert header.headerData(1, Qt.Horizontal,
Qt.DisplayRole) == "b"
assert header.headerData(4, Qt.Horizontal,
Qt.DisplayRole) == "e"
def test_non_ascii_index():
"""
Test that there are no errors when displaying a dataframe with
a non-ascii index and header.
"""
df = read_csv(os.path.join(FILES_PATH, 'issue_5833.csv'), index_col=0)
editor = DataFrameEditor(None)
editor.setup_and_check(df)
index = editor.table_index.model()
header = editor.table_header.model()
dfm = editor.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == "кодирование"
assert data_index(index, 0, 0) == 'пример'
assert data(dfm, 0, 0) == 'файла'
def test_no_convert_strings_to_unicode():
"""
Test that we don't apply any conversion to strings in headers,
indexes or data.
"""
df = read_csv(os.path.join(FILES_PATH, 'issue_5833.csv'), index_col=0,
encoding='koi8_r')
editor = DataFrameEditor(None)
editor.setup_and_check(df)
index = editor.table_index.model()
header = editor.table_header.model()
dfm = editor.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) != u"кодирование"
assert data_index(index, 0, 0) != u'пример'
assert data(dfm, 0, 0) != u'файла'
def test_header_encoding():
"""Test for header encoding handling."""
df = read_csv(os.path.join(FILES_PATH, 'issue_3896.csv'))
editor = DataFrameEditor(None)
editor.setup_and_check(df)
header = editor.table_header.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == "Unnamed: 0"
assert "Unieke_Idcode" in header.headerData(1, Qt.Horizontal,
Qt.DisplayRole)
assert header.headerData(2, Qt.Horizontal,
Qt.DisplayRole) == "a"
assert header.headerData(3, Qt.Horizontal,
Qt.DisplayRole) == "b"
assert header.headerData(4, Qt.Horizontal,
Qt.DisplayRole) == "c"
assert header.headerData(5, Qt.Horizontal,
Qt.DisplayRole) == "d"
def test_dataframeeditor_with_various_indexes():
for rng_name, rng in generate_pandas_indexes().items():
editor = DataFrameEditor(None)
editor.setup_and_check(rng)
dfm = editor.dataModel
assert dfm.rowCount() == 20
assert dfm.columnCount() == 1
header = editor.table_header.model()
assert header.headerData(0, Qt.Horizontal,
Qt.DisplayRole) == "0"
if rng_name == "Index":
assert data(dfm, 0, 0) == 'A'
assert data(dfm, 1, 0) == 'B'
assert data(dfm, 2, 0) == 'C'
assert data(dfm, 19, 0) == 'T'
elif rng_name == "RangeIndex":
assert data(dfm, 0, 0) == '0'
assert data(dfm, 1, 0) == '1'
assert data(dfm, 2, 0) == '2'
assert data(dfm, 19, 0) == '19'
elif rng_name == "Float64Index":
assert data(dfm, 0, 0) == '0'
assert data(dfm, 1, 0) == '0.1'
assert data(dfm, 2, 0) == '0.2'
assert data(dfm, 19, 0) == '1.9'
elif rng_name == "DatetimeIndex":
assert data(dfm, 0, 0) == '2017-01-01 00:00:00'
assert data(dfm, 1, 0) == '2017-01-02 00:00:00'
assert data(dfm, 2, 0) == '2017-01-03 00:00:00'
assert data(dfm, 19, 0) == '2017-01-20 00:00:00'
elif rng_name == "MultiIndex":
assert data(dfm, 0, 0) == "('A', 'foo')"
assert data(dfm, 1, 0) == "('A', 'bar')"
assert data(dfm, 2, 0) == "('B', 'foo')"
assert data(dfm, 19, 0) == "('J', 'bar')"
elif rng_name == "CategoricalIndex":
assert data(dfm, 0, 0) == 'a'
assert data(dfm, 1, 0) == 'b'
assert data(dfm, 2, 0) == 'c'
assert data(dfm, 19, 0) == 'b'