def test_display_float_scientific_a(self) -> None:
s1 = sf.Series([3.1, 5.2])**40
self.assertEqual(
s1.display(sf.DisplayConfig(type_color=False)).to_rows(), [
'<Series>', '<Index>', '0 4.51302515e+19',
'1 4.36650282e+28', '<int64> <float64>'
])
# non default config for scientifici will truncate values
self.assertEqual(
s1.display(
sf.DisplayConfig(
type_color=False,
value_format_float_scientific='{:f}')).to_rows(), [
'<Series>', '<Index>', '0 45130251461102272...',
'1 43665028242109266...', '<int64> <float64>'
])
self.assertEqual(
s1.display(
sf.DisplayConfig(
type_color=False,
value_format_float_scientific='{:.2e}')).to_rows(), [
'<Series>', '<Index>', '0 4.51e+19',
'1 4.37e+28', '<int64> <float64>'
])
def test_display_float_scientific_b(self) -> None:
s1 = sf.Series([3.1j, 5.2j])**40
self.assertEqual(
s1.display(sf.DisplayConfig(type_color=False)).to_rows(), [
'<Series>', '<Index>', '0 4.51e+19+0.00e+00j',
'1 4.37e+28+0.00e+00j', '<int64> <complex128>'
])
# non default config for scientifici will truncate values
self.assertEqual(
s1.display(
sf.DisplayConfig(
type_color=False,
value_format_complex_scientific='{:f}')).to_rows(),
[
'<Series>', '<Index>', '0 45130251461102338...',
'1 43665028242109283...', '<int64> <complex128>'
])
self.assertEqual(
s1.display(
sf.DisplayConfig(
type_color=False,
value_format_complex_scientific='{:.1e}')).to_rows(), [
'<Series>', '<Index>', '0 4.5e+19+0.0e+00j',
'1 4.4e+28+0.0e+00j', '<int64> <complex128>'
])
def test_display_html_index_b(self) -> None:
records = (
(2, 'a', False),
(30, 'b', False),
)
f = Frame.from_records(records,
columns=('p', 'q', 'r'),
index=('w', 'x'))
f = f.set_index_hierarchy(('p', 'q'))
# this uses cell width normaliz
post = f.display(
sf.DisplayConfig(display_format=DisplayFormats.HTML_PRE,
cell_max_width_leftmost=20)).to_rows()
self.assertEqual(
post[2],
'<span style="color: #777777"><IndexHierarchy: ...</span>')
post = f.display(
sf.DisplayConfig(display_format=DisplayFormats.HTML_PRE,
cell_max_width_leftmost=36)).to_rows()
self.assertEqual(
post[2],
'<span style="color: #777777"><IndexHierarchy: ('p', 'q')></span>'
)
def test_display_html_table_a(self) -> None:
f = sf.Frame.from_dict(
dict(a=(1, 2, 3, 4), b=(True, False, True, False), c=list('qrst')))
f = f.set_index_hierarchy(['a', 'b'])
f = f.relabel_add_level(columns='I')
f = f.relabel_add_level(columns='J')
expected1 = f.display(
sf.DisplayConfig(display_format='html_table', type_color=False))
html1 = '''<table border="1"><thead><tr><th><Frame></th><th></th><th></th><th></th><th></th><th></th></tr><tr><th><IndexHierarchy></th><th></th><th>J</th><th>J</th><th>J</th><th><<U1></th></tr><tr><th></th><th></th><th>I</th><th>I</th><th>I</th><th><<U1></th></tr><tr><th></th><th></th><th>a</th><th>b</th><th>c</th><th><<U1></th></tr><tr><th><IndexHierarchy: ('a', 'b')></th><th></th><th></th><th></th><th></th><th></th></tr></thead><tbody><tr><th>1</th><th>True</th><td>1</td><td>True</td><td>q</td><td></td></tr><tr><th>2</th><th>False</th><td>2</td><td>False</td><td>r</td><td></td></tr><tr><th>3</th><th>True</th><td>3</td><td>True</td><td>s</td><td></td></tr><tr><th>4</th><th>False</th><td>4</td><td>False</td><td>t</td><td></td></tr><tr><th><int64></th><th><bool></th><td><int64></td><td><bool></td><td><<U1></td><td></td></tr></tbody></table>
'''
self.assertEqual(html1.strip(), str(expected1).strip())
expected2 = f.display(
sf.DisplayConfig(display_format='html_table',
type_color=False,
type_show=False))
html2 = '''<table border="1"><thead><tr><th></th><th></th><th>J</th><th>J</th><th>J</th></tr><tr><th></th><th></th><th>I</th><th>I</th><th>I</th></tr><tr><th></th><th></th><th>a</th><th>b</th><th>c</th></tr></thead><tbody><tr><th>1</th><th>True</th><td>1</td><td>True</td><td>q</td></tr><tr><th>2</th><th>False</th><td>2</td><td>False</td><td>r</td></tr><tr><th>3</th><th>True</th><td>3</td><td>True</td><td>s</td></tr><tr><th>4</th><th>False</th><td>4</td><td>False</td><td>t</td></tr></tbody></table>
'''
self.assertEqual(html2.strip(), str(expected2).strip())
expected3 = f.display(
sf.DisplayConfig(display_format='html_table',
type_color=False,
type_show=False,
include_index=False))
html3 = '''<table border="1"><thead><tr><th>J</th><th>J</th><th>J</th></tr><tr><th>I</th><th>I</th><th>I</th></tr><tr><th>a</th><th>b</th><th>c</th></tr></thead><tbody><tr><td>1</td><td>True</td><td>q</td></tr><tr><td>2</td><td>False</td><td>r</td></tr><tr><td>3</td><td>True</td><td>s</td></tr><tr><td>4</td><td>False</td><td>t</td></tr></tbody></table>
'''
self.assertEqual(html3.strip(), str(expected3).strip())
expected4 = f.display(
sf.DisplayConfig(display_format='html_table',
type_color=False,
type_show=False,
include_columns=False))
html4 = '''<table border="1"><tbody><tr><th>1</th><th>True</th><td>1</td><td>True</td><td>q</td></tr><tr><th>2</th><th>False</th><td>2</td><td>False</td><td>r</td></tr><tr><th>3</th><th>True</th><td>3</td><td>True</td><td>s</td></tr><tr><th>4</th><th>False</th><td>4</td><td>False</td><td>t</td></tr></tbody></table>
'''
self.assertEqual(html4.strip(), str(expected4).strip())
expected5 = f.display(
sf.DisplayConfig(display_format='html_table',
type_color=False,
type_show=False,
include_index=False,
include_columns=False))
html5 = '''<table border="1"><tbody><tr><td>1</td><td>True</td><td>q</td></tr><tr><td>2</td><td>False</td><td>r</td></tr><tr><td>3</td><td>True</td><td>s</td></tr><tr><td>4</td><td>False</td><td>t</td></tr></tbody></table>
'''
self.assertEqual(html5.strip(), str(expected5).strip())
def run_test():
records = []
for label, fixture in (('None', None), ):
for cls in (
MemorySum,
MemoryThreadSum,
MemoryForkSum,
MemorySpawnSum,
MMapThreadSum,
MMapForkSum,
MMapSpawnSum,
):
runner = cls(fixture)
record = [cls.__name__, cls.NUMBER, label]
result = timeit.timeit(f'runner()',
globals=locals(),
number=cls.NUMBER)
record.append(result)
records.append(record)
runner.clear()
f = sf.FrameGO.from_records(records,
columns=('name', 'number', 'fixture', 'time'))
display = f.iter_element_items().apply(get_format())
config = sf.DisplayConfig(
cell_max_width_leftmost=np.inf,
cell_max_width=np.inf,
type_show=False,
display_rows=200,
include_index=False,
)
print(display.display(config))
def test_display_max_width_a(self) -> None:
records = (
(2, 'a', False),
(30, 'b', False),
)
f1 = Frame.from_records(records,
columns=('p', 'q', 'r'),
index=('w', 'x'))
f2 = f1.set_index_hierarchy(('p', 'q'))
post = f2.display(
sf.DisplayConfig(type_color=False,
cell_max_width_leftmost=20)).to_rows()
self.assertEqual(post[2], '<IndexHierarchy: ...')
post = f2.display(
sf.DisplayConfig(type_color=False,
cell_max_width_leftmost=30)).to_rows()
self.assertEqual(post[2], "<IndexHierarchy: ('p', 'q')>")
def test_display_html_pre_a(self):
f = Frame(dict(a=(1, 2), b=(1.2, 3.4), c=(False, True)))
expected = f.display(
sf.DisplayConfig(display_format='html_pre', type_color=False))
html = '''<div style="white-space: pre; font-family: monospace"><Frame>
<Index> a b c <<U1>
<Index>
0 1 1.2 False
1 2 3.4 True
<int64> <int64> <float64> <bool></div>'''
self.assertEqual(html.strip(), str(expected).strip())
def test_display_html_table_a(self) -> None:
f = sf.Frame.from_dict(
dict(a=(1, 2, 3, 4), b=(True, False, True, False), c=list('qrst')))
f = f.set_index_hierarchy(['a', 'b'])
f = f.relabel_add_level(columns='I')
f = f.relabel_add_level(columns='J')
expected = f.display(
sf.DisplayConfig(display_format='html_table', type_color=False))
html = '''
<table border="1"><thead><tr><th><Frame></th><th></th><th></th><th></th><th></th><th></th></tr><tr><th><IndexHierarchy></th><th></th><th>J</th><th>J</th><th>J</th><th><<U1></th></tr><tr><th></th><th></th><th>I</th><th>I</th><th>I</th><th><<U1></th></tr><tr><th></th><th></th><th>a</th><th>b</th><th>c</th><th><<U1></th></tr><tr><th><IndexHierarchy></th><th></th><th></th><th></th><th></th><th></th></tr></thead><tbody><tr><th>1</th><th>True</th><td>1</td><td>True</td><td>q</td><td></td></tr><tr><th>2</th><th>False</th><td>2</td><td>False</td><td>r</td><td></td></tr><tr><th>3</th><th>True</th><td>3</td><td>True</td><td>s</td><td></td></tr><tr><th>4</th><th>False</th><td>4</td><td>False</td><td>t</td><td></td></tr><tr><th><object></th><th><object></th><td><int64></td><td><bool></td><td><<U1></td><td></td></tr></tbody></table>
'''
self.assertEqual(html.strip(), str(expected).strip())
def grammar(context):
from frame_fixtures.core import GrammarDoc
import static_frame as sf #type: ignore
config = sf.DisplayConfig(include_index=False,
type_show=False,
cell_max_width=40)
# container components
cc = GrammarDoc.container_components()
print(cc.to_rst(config))
# container types
cc = GrammarDoc.specifiers_constructor()
print(cc.to_rst(config))
# container types
cc = GrammarDoc.specifiers_dtype()
print(cc.to_rst(config))
def run_test():
records = []
for cls in (
DirectFromString,
ParseFromString,
ShiftDay,
DeriveYear,
TrueOnMonday,
TrueOnEOM,
TrueLessThan,
ConvertToStr,
):
runner = cls()
record = [cls.__name__, cls.NUMBER]
for func_name in ('dt', 'dt64'):
result = timeit.timeit(
f'runner.{func_name}()',
globals=locals(),
number=cls.NUMBER)
record.append(result)
records.append(record)
f = sf.FrameGO.from_records(records,
columns=('name', 'number', 'dt', 'dt64')
)
f['dt64/dt'] = f['dt64'] / f['dt']
f['dt/dt64'] = f['dt'] / f['dt64']
f['dt64-faster'] = f['dt/dt64'] > 1
display = f.iter_element_items().apply(get_format())
config = sf.DisplayConfig(
cell_max_width_leftmost=np.inf,
cell_max_width=np.inf,
type_show=False,
display_rows=200,
include_index=False,
)
print(display.display(config))
def run_test():
records = []
for label, fixture in (
('wide', FF_wide),
('tall', FF_tall),
('square', FF_square),
):
# for label, fixture in (('square', FF_square),):
fixtures = [fixture] * 100
for cls in (
FileWriteBusParquet,
FileWriteBusNPZ,
# FileWritePickle,
FileReadBusParquet,
FileReadBusNPZ,
# FileReadPickle,
):
runner = cls(fixtures)
record = [cls.__name__, cls.NUMBER, label]
result = timeit.timeit(f'runner()',
globals=locals(),
number=cls.NUMBER)
record.append(result)
records.append(record)
f = sf.FrameGO.from_records(records,
columns=('name', 'number', 'fixture', 'time'))
display = f.iter_element_items().apply(get_format())
config = sf.DisplayConfig(
cell_max_width_leftmost=np.inf,
cell_max_width=np.inf,
type_show=False,
display_rows=200,
include_index=False,
)
print(display.display(config))
ThreadPool,
# ProcessPool,
):
runner = cls(size, groups)
record = [cls.__name__, cls.NUMBER, size, groups]
result = timeit.timeit(f'runner.run()',
globals=locals(),
number=cls.NUMBER)
record.append(result)
records.append(record)
f = sf.FrameGO.from_records(records,
columns=('name', 'number', 'size', 'groups',
'time'))
display = f.iter_element_items().apply(get_format())
config = sf.DisplayConfig(
cell_max_width_leftmost=np.inf,
cell_max_width=np.inf,
type_show=False,
display_rows=200,
include_index=False,
)
print(display.display(config))
# import ipdb; ipdb.set_trace()
if __name__ == '__main__':
run_test()
def run_test():
records = []
for dtype_hetero, fixture_label, fixture in (
fixture_to_pair('uniform', FF_wide_uniform),
fixture_to_pair('mixed', FF_wide_mixed),
fixture_to_pair('columnar', FF_wide_columnar),
fixture_to_pair('uniform', FF_tall_uniform),
fixture_to_pair('mixed', FF_tall_mixed),
fixture_to_pair('columnar', FF_tall_columnar),
fixture_to_pair('uniform', FF_square_uniform),
fixture_to_pair('mixed', FF_square_mixed),
fixture_to_pair('columnar', FF_square_columnar),
):
cls_read = (
PDReadParquetArrow,
PDReadParquetArrowNoComp,
# PDReadParquetFast, # not faster!
PDReadFeather,
# SFReadParquet,
SFReadNPZ,
SFReadNPY,
SFReadPickle,
# SFReadNPYMM,
)
cls_write = (
PDWriteParquetArrow,
PDWriteParquetArrowNoComp,
# PDWriteParquetFast, # not faster!
# SFWriteParquet,
PDWriteFeather,
SFWriteNPZ,
SFWriteNPY,
SFWritePickle,
)
for cls, category_prefix in chain(
zip(cls_read, repeat('read')),
zip(cls_write, repeat('write')),
):
runner = cls(fixture)
category = f'{category_prefix} {dtype_hetero}'
record = [cls.__name__, cls.NUMBER, category, fixture_label]
try:
result = timeit.timeit(f'runner()',
globals=locals(),
number=cls.NUMBER)
except OSError:
result = np.nan
finally:
runner.clear()
record.append(result)
records.append(record)
f = sf.FrameGO.from_records(records,
columns=('name', 'number', 'category',
'fixture', 'time'))
display = f.iter_element_items().apply(get_format())
config = sf.DisplayConfig(
cell_max_width_leftmost=np.inf,
cell_max_width=np.inf,
type_show=False,
display_rows=200,
include_index=False,
)
print(display.display(config))
plot(f)
def tables() -> None:
name = 'Higher-Order Containers'
columns = ('Bus', 'Batch', 'Quilt', 'Yarn')
records_items = (
('Presented Dimensions', (
1,
2,
2,
1,
)),
('Approximate Interface', (
'Series',
'Frame',
'Frame',
'Series',
)),
('Composes', (
'n Frame',
'',
'1 Bus or Yarn of n Frame',
'm Bus of n Frame',
)),
('Presented shape', (
'(n,)',
'',
'(xn, y) or (x, yn)',
'(mn,)',
)),
)
f = sf.Frame.from_records_items(records_items, columns=columns, name=name)
# print(f)
config = sf.DisplayConfig(cell_max_width=200, type_show=False)
print(f.name)
print(f.to_rst(config=config))
name = 'Constructors & Exporters'
columns = ('Constructor', 'Exporter') #type: ignore
records = (
('from_hdf5', 'to_hdf5'),
('from_sqlite', 'to_sqlite'),
(
'from_zip_csv',
'to_zip_csv',
),
(
'from_zip_npz',
'to_zip_npz',
),
(
'from_zip_pickle',
'to_zip_pickle',
),
(
'from_zip_parquet',
'to_zip_parquet',
),
(
'from_zip_tsv',
'to_zip_tsv',
),
('from_xlsx', 'to_xlsx'),
)
f = sf.Frame.from_records(records, columns=columns, name=name)
# print(f)
print(f.name)
print(f.to_rst(sf.DisplayConfig(include_index=False, type_show=False)))