def test_recursive_rowfunc():
f = rrowfunc(t['name'], t)
assert [f(row) for row in data] == [row[0] for row in data]
f = rrowfunc(t['amount'] + t['id'], t)
assert [f(row) for row in data] == [row[1] + row[2] for row in data]
assert raises(Exception, lambda: rrowfunc(t[t['amount'] < 0]['name'], t))
def test_recursive_rowfunc():
f = rrowfunc(t['name'], t)
assert [f(row) for row in data] == [row[0] for row in data]
f = rrowfunc(t['amount'] + t['id'], t)
assert [f(row) for row in data] == [row[1] + row[2] for row in data]
assert raises(Exception, lambda: rrowfunc(t[t['amount'] < 0]['name'], t))
def test_gzopen_json():
with tmpfile('.json.gz') as filename:
with gzip.open(filename, 'w') as f:
f.write('[[1, 1], [2, 2]]')
# Not a valid JSON file
assert raises(Exception, lambda: list(JSON(filename, schema='2 * int')))
dd = JSON(filename, schema='2 * int', open=gzip.open)
assert list(dd) == [[1, 1], [2, 2]]
def test_gzopen_csv():
with tmpfile('.csv.gz') as filename:
with gzip.open(filename, 'w') as f:
f.write('1,1\n2,2')
# Not a valid CSV file
assert raises(Exception, lambda: list(CSV(filename, schema='2 * int')))
dd = CSV(filename, schema='2 * int', open=gzip.open)
assert list(dd) == [[1, 1], [2, 2]]
def test_union():
schema = '{x: int, y: int, z: int}'
a = TableSymbol('a', schema)
b = TableSymbol('b', schema)
c = TableSymbol('c', schema)
u = union(a, b, c)
assert u.schema == a.schema
assert raises(Exception,
lambda: union(a, TableSymbol('q', '{name: string}')))
def test_gzopen_csv():
with tmpfile('.csv.gz') as filename:
with gzip.open(filename, 'w') as f:
f.write('1,1\n2,2')
# Not a valid CSV file
assert raises(Exception, lambda: list(CSV(filename, schema='2 * int')))
dd = CSV(filename, schema='2 * int', open=gzip.open)
assert list(dd) == [[1, 1], [2, 2]]
def test_union():
schema = '{x: int, y: int, z: int}'
a = TableSymbol('a', schema)
b = TableSymbol('b', schema)
c = TableSymbol('c', schema)
u = union(a, b, c)
assert u.schema == a.schema
assert raises(Exception,
lambda: union(a, TableSymbol('q', '{name: string}')))
def test_gzopen_json():
with tmpfile('.json.gz') as filename:
with gzip.open(filename, 'w') as f:
f.write('[[1, 1], [2, 2]]')
# Not a valid JSON file
assert raises(Exception,
lambda: list(JSON(filename, schema='2 * int')))
dd = JSON(filename, schema='2 * int', open=gzip.open)
assert list(dd) == [[1, 1], [2, 2]]
def test_gzopen_json():
with tmpfile('.json.gz') as filename:
f = gzip.open(filename, 'wt')
f.write('[[1, 1], [2, 2]]')
f.close()
# Not a valid JSON file
assert raises(Exception, lambda: list(JSON(filename, schema='2 * int')))
dd = JSON(filename, schema='2 * int', open=gzip.open)
assert tuplify(list(dd)) == ((1, 1), (2, 2))
def test_table_creation(self):
dd = SQL(self.engine,
'testtable',
schema='{name: string, amount: int}',
primary_key='name')
assert self.engine.has_table('testtable')
assert dd.table.columns.get('name').primary_key
assert not dd.table.columns.get('amount').primary_key
assert dd.dshape == dshape('var * {name: string, amount: int}')
assert raises(ValueError, lambda: SQL(self.engine, 'testtable2'))
def test_table_creation(self):
dd = SQL(self.engine, 'testtable',
schema='{name: string, amount: int}',
primary_key='name')
assert self.engine.has_table('testtable')
assert dd.table.columns.get('name').primary_key
assert not dd.table.columns.get('amount').primary_key
assert dd.dshape == dshape('var * {name: string, amount: int}')
assert raises(ValueError, lambda: SQL(self.engine, 'testtable2'))
def test_gzopen_json():
with tmpfile('.json.gz') as filename:
f = gzip.open(filename, 'wt')
f.write('[[1, 1], [2, 2]]')
f.close()
# Not a valid JSON file
assert raises(Exception,
lambda: list(JSON(filename, schema='2 * int')))
dd = JSON(filename, schema='2 * int', open=gzip.open)
assert tuplify(list(dd)) == ((1, 1), (2, 2))
def test_gzopen_csv():
with tmpfile('.csv.gz') as filename:
f = gzip.open(filename, 'wt')
f.write('1,1\n2,2')
f.close()
# Not a valid CSV file
assert raises(Exception, lambda: list(CSV(filename, schema='2 * int')))
dd = CSV(filename, schema='2 * int', open=partial(gzip.open, mode='rt'))
assert tuplify(list(dd)) == ((1, 1), (2, 2))
def test_selection_typecheck():
t = symbol('t', 'var * {name: string, amount: int, id: int}')
assert raises(TypeError, lambda: t[t['amount'] + t['id']])
assert raises(TypeError, lambda: t[t['name']])
def test_raise_error_if_join_on_no_columns():
a = symbol('a', 'var * {x: int}')
b = symbol('b', 'var * {y: int}')
assert raises(ValueError, lambda: join(a, b))
def test_multiple_renames_on_series_fails():
t = symbol('s', 'var * {timestamp: datetime}')
ts = t.timestamp
assert raises(ValueError, lambda: ts.relabel({'timestamp': 'date',
'hello': 'world'}))
def test_map_with_rename():
t = symbol('s', 'var * {timestamp: datetime}')
result = t.timestamp.map(lambda x: x.date(), schema='{date: datetime}')
assert raises(ValueError, lambda: result.relabel({'timestamp': 'date'}))
assert result.fields == ['date']
from blaze import CSV, JSON
import subprocess
import tempfile
import json
import os
from blaze.utils import filetext, tmpfile, raises
from blaze.compatibility import PY3
from datashape import discover, dshape
from blaze import drop, into, create_index
from blaze.utils import assert_allclose
no_mongoimport = pytest.mark.skipif(raises(OSError,
lambda : subprocess.Popen('mongoimport',
shell=os.name != 'nt',
stdout=subprocess.PIPE).wait()),
reason='mongoimport cannot be found')
@pytest.yield_fixture(scope='module')
def conn():
pymongo = pytest.importorskip('pymongo')
try:
c = pymongo.MongoClient()
except pymongo.errors.ConnectionFailure:
pytest.skip('No mongo server running')
else:
yield c
c.close()
def test_shapes_raise_errors():
assert raises(ValueError, lambda: w + x)
def test_improper_selection():
t = TableSymbol('t', '{x: int, y: int, z: int}')
assert raises(Exception, lambda: t[t['x'] > 0][t.sort()[t['y' > 0]]])
def test_raise_error_if_join_on_no_columns():
a = symbol('a', 'var * {x: int}')
b = symbol('b', 'var * {y: int}')
assert raises(ValueError, lambda: join(a, b))
def test_improper_selection():
t = symbol('t', 'var * {x: int, y: int, z: int}')
assert raises(Exception, lambda: t[t['x'] > 0][t.sort()[t['y' > 0]]])
from blaze.data import CSV, JSON
import subprocess
import tempfile
import json
import os
from blaze.utils import filetext, tmpfile, raises
from blaze.compatibility import PY3, PY2
from datashape import discover, dshape
from blaze import drop, into, create_index
from blaze.utils import assert_allclose
from blaze.resource import resource
no_mongoimport = pytest.mark.skipif(raises(
OSError, lambda: subprocess.Popen(
'mongoimport', shell=os.name != 'nt', stdout=subprocess.PIPE).wait()),
reason='mongoimport cannot be found')
@pytest.yield_fixture(scope='module')
def conn():
pymongo = pytest.importorskip('pymongo')
try:
c = pymongo.MongoClient()
except pymongo.errors.ConnectionFailure:
pytest.skip('No mongo server running')
else:
yield c
c.close()
def test_iter_raises_not_implemented_Error():
e = Symbol('e', '5 * {x: int, "a b": int}')
assert raises(NotImplementedError, lambda: iter(e))
def test_shapes_raise_errors():
assert raises(ValueError, lambda: w + x)
def test_dir():
i = symbol('i', '10 * int')
d = symbol('d', '10 * datetime')
assert i + 1
assert raises(Exception, lambda: d + 1)
def test_improper_selection():
t = TableSymbol("t", "{x: int, y: int, z: int}")
assert raises(Exception, lambda: t[t["x"] > 0][t.sort()[t["y" > 0]]])
from blaze.utils import raises
from odo import URL, CSV
import pandas as pd
import pandas.util.testing as tm
from functools import partial
try:
from urllib2 import urlopen
from urllib2 import HTTPError, URLError
except ImportError:
from urllib.request import urlopen
from urllib.error import HTTPError, URLError
pytestmark = pytest.mark.skipif(raises(URLError,
partial(urlopen, "http://google.com")),
reason='unable to connect to google.com')
iris_url = ('https://raw.githubusercontent.com/'
'blaze/blaze/master/blaze/examples/data/iris.csv')
@pytest.fixture
def iris_local():
thisdir = os.path.abspath(os.path.dirname(__file__))
return data(os.path.join(thisdir, os.pardir, os.pardir, "examples", "data", "iris.csv"))
def test_url_csv_data(iris_local):
iris_remote = data(iris_url)
assert isinstance(iris_remote.data, URL(CSV))
iris_remote_df = compute(iris_remote)
assert isinstance(iris_remote_df, pd.DataFrame)
def test_selection_typecheck():
t = TableSymbol('t', '{name: string, amount: int, id: int}')
assert raises(TypeError, lambda: t[t['amount'] + t['id']])
assert raises(TypeError, lambda: t[t['name']])
from blaze.utils import raises
from odo import URL, CSV
import pandas as pd
import pandas.util.testing as tm
from functools import partial
try:
from urllib2 import urlopen
from urllib2 import HTTPError, URLError
except ImportError:
from urllib.request import urlopen
from urllib.error import HTTPError, URLError
pytestmark = pytest.mark.skipif(raises(URLError,
partial(urlopen, "http://google.com")),
reason='unable to connect to google.com')
iris_url = ('https://raw.githubusercontent.com/'
'blaze/blaze/master/blaze/examples/data/iris.csv')
@pytest.fixture
def iris_local():
thisdir = os.path.abspath(os.path.dirname(__file__))
return Data(
os.path.join(thisdir, os.pardir, os.pardir, "examples", "data",
"iris.csv"))
def test_url_csv_data(iris_local):
def test_selection_typecheck():
t = TableSymbol("t", "{name: string, amount: int, id: int}")
assert raises(TypeError, lambda: t[t["amount"] + t["id"]])
assert raises(TypeError, lambda: t[t["name"]])
def test_iter_raises_not_implemented_Error():
e = symbol('e', '5 * {x: int, "a b": int}')
assert raises(NotImplementedError, lambda: iter(e))