def test_native_iterator():
# Get table
def generator():
yield ['id', 'name']
yield ['1', 'english']
yield ['2', '中国人']
with pytest.raises(exceptions.ParsingError) as excinfo:
iterator = generator()
topen(iterator)
assert 'callable' in str(excinfo.value)
def test_storage():
# Get resources
articles_schema = json.load(io.open('data/articles.json', encoding='utf-8'))
comments_schema = json.load(io.open('data/comments.json', encoding='utf-8'))
articles_data = topen('data/articles.csv', with_headers=True).read()
comments_data = topen('data/comments.csv', with_headers=True).read()
# Engine
engine = create_engine(os.environ['DATABASE_URL'])
# Storage
storage = Storage(engine=engine, prefix='prefix_')
# Delete tables
for table in reversed(storage.tables):
storage.delete(table)
# Create tables
storage.create(['articles', 'comments'], [articles_schema, comments_schema])
# Write data to tables
storage.write('articles', articles_data)
storage.write('comments', comments_data)
# Create new storage to use reflection only
storage = Storage(engine=engine, prefix='prefix_')
# Create existent table
with pytest.raises(RuntimeError):
storage.create('articles', articles_schema)
# Get table representation
assert repr(storage).startswith('Storage')
# Get tables list
assert storage.tables == ['articles', 'comments']
# Get table schemas
assert storage.describe('articles') == convert_schema(articles_schema)
assert storage.describe('comments') == convert_schema(comments_schema)
# Get table data
assert list(storage.read('articles')) == convert_data(articles_schema, articles_data)
assert list(storage.read('comments')) == convert_data(comments_schema, comments_data)
# Delete tables
for table in reversed(storage.tables):
storage.delete(table)
# Delete non existent table
with pytest.raises(RuntimeError):
storage.delete('articles')
def test_native_iterator():
# Get table
def generator():
yield ['id', 'name']
yield ['1', 'english']
yield ['2', '中国人']
with pytest.raises(exceptions.ParsingError) as excinfo:
iterator = generator()
topen(iterator)
assert 'callable' in str(excinfo.value)
def test_native_iterator():
# Get table
def generator():
yield ["id", "name"]
yield ["1", "english"]
yield ["2", "中国人"]
with pytest.raises(exceptions.SourceError) as excinfo:
iterator = generator()
topen(iterator)
assert "callable" in str(excinfo.value)
def test_save_csv(tmpdir):
# Save table
path = str(tmpdir.join("table.csv"))
table = topen("data/table.csv", headers=1)
table.save(path)
# Open saved table
table = topen(path, headers=1)
# Make assertions
assert table.headers == ["id", "name"]
assert table.read(extended=True) == [(2, ["id", "name"], ["1", "english"]), (3, ["id", "name"], ["2", "中国人"])]
def test_save_csv(tmpdir):
# Save table
path = str(tmpdir.join('table.csv'))
table = topen('data/table.csv', headers=1)
table.save(path)
# Open saved table
table = topen(path, headers=1)
# Make assertions
assert table.headers == ['id', 'name']
assert table.read(extended=True) == [(2, ['id', 'name'], ['1', 'english']),
(3, ['id', 'name'], ['2', '中国人'])]
def test_save_csv(tmpdir):
# Save table
path = str(tmpdir.join('table.csv'))
table = topen('data/table.csv', headers=1)
table.save(path)
# Open saved table
table = topen(path, headers=1)
# Make assertions
assert table.headers == ['id', 'name']
assert table.read(extended=True) == [
(2, ['id', 'name'], ['1', 'english']),
(3, ['id', 'name'], ['2', '中国人'])]
def test_file_csv_with_bom():
# Get table
table = topen("data/special/bom.csv", encoding="utf-8")
# Make assertions
assert table.headers is None
assert table.read() == [["id", "name"], ["1", "english"], ["2", "中国人"]]
# Get table
table = topen("data/special/bom.csv")
# Make assertions
assert table.headers is None
assert table.read() == [["id", "name"], ["1", "english"], ["2", "中国人"]]
def test_file_csv_with_bom():
# Get table
table = topen('data/special/bom.csv', encoding='utf-8')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
# Get table
table = topen('data/special/bom.csv')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_file_csv_with_bom():
# Get table
table = topen('data/special/bom.csv', encoding='utf-8')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
# Get table
table = topen('data/special/bom.csv')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_html_content():
# Check raises
source = 'https://github.com/frictionlessdata/tabulator-py/blob/master/data/table.csv'
with pytest.raises(exceptions.TabulatorException) as excinfo:
table = topen(source, headers='row1')
assert 'HTML' in str(excinfo.value)
def test_processors_chain():
# Processors
def skip_commented_rows(extended_rows):
for number, headers, row in extended_rows:
if row and hasattr(row[0], "startswith") and row[0].startswith("#"):
continue
yield (number, headers, row)
def skip_blank_rows(extended_rows):
for number, headers, row in extended_rows:
if not row:
continue
yield (number, headers, row)
def cast_rows(extended_rows):
for number, headers, row in extended_rows:
crow = []
for value in row:
try:
if isinstance(value, six.string_types):
value = ast.literal_eval(value)
except Exception:
pass
crow.append(value)
yield (number, headers, crow)
# Get table
source = [["id", "name"], ["#1", "english"], [], ["2", "中国人"]]
table = topen(source, headers="row1", post_parse=[skip_commented_rows, skip_blank_rows, cast_rows])
# Make assertions
assert table.headers == ["id", "name"]
def test_reset_and_sample_size():
# Get table
table = topen("data/special/long.csv", headers=1, sample_size=3)
# Make assertions
assert table.read(extended=True) == [
(2, ["id", "name"], ["1", "a"]),
(3, ["id", "name"], ["2", "b"]),
(4, ["id", "name"], ["3", "c"]),
(5, ["id", "name"], ["4", "d"]),
(6, ["id", "name"], ["5", "e"]),
(7, ["id", "name"], ["6", "f"]),
]
assert table.sample == [["1", "a"], ["2", "b"]]
assert table.read() == []
# Reset table
table.reset()
# Make assertions
assert table.read(extended=True, limit=3) == [
(2, ["id", "name"], ["1", "a"]),
(3, ["id", "name"], ["2", "b"]),
(4, ["id", "name"], ["3", "c"]),
]
assert table.sample == [["1", "a"], ["2", "b"]]
assert table.read(extended=True) == [
(5, ["id", "name"], ["4", "d"]),
(6, ["id", "name"], ["5", "e"]),
(7, ["id", "name"], ["6", "f"]),
]
def test_html_content():
# Check raises
source = "https://github.com/frictionlessdata/tabulator-py/blob/master/data/table.csv"
with pytest.raises(exceptions.FormatError) as excinfo:
table = topen(source, headers="row1")
assert "HTML" in str(excinfo.value)
def test_reset_and_sample_size():
# Get table
table = topen('data/special/long.csv', headers=1, sample_size=3)
# Make assertions
assert table.read(extended=True) == [(2, ['id', 'name'], ['1', 'a']),
(3, ['id', 'name'], ['2', 'b']),
(4, ['id', 'name'], ['3', 'c']),
(5, ['id', 'name'], ['4', 'd']),
(6, ['id', 'name'], ['5', 'e']),
(7, ['id', 'name'], ['6', 'f'])]
assert table.sample == [['1', 'a'], ['2', 'b']]
assert table.read() == []
# Reset table
table.reset()
# Make assertions
assert table.read(extended=True,
limit=3) == [(2, ['id', 'name'], ['1', 'a']),
(3, ['id', 'name'], ['2', 'b']),
(4, ['id', 'name'], ['3', 'c'])]
assert table.sample == [['1', 'a'], ['2', 'b']]
assert table.read(extended=True) == [(5, ['id', 'name'], ['4', 'd']),
(6, ['id', 'name'], ['5', 'e']),
(7, ['id', 'name'], ['6', 'f'])]
def test_reset_and_sample_size():
# Get table
table = topen('data/special/long.csv', headers=1, sample_size=3)
# Make assertions
assert table.read(extended=True) == [
(2, ['id', 'name'], ['1', 'a']),
(3, ['id', 'name'], ['2', 'b']),
(4, ['id', 'name'], ['3', 'c']),
(5, ['id', 'name'], ['4', 'd']),
(6, ['id', 'name'], ['5', 'e']),
(7, ['id', 'name'], ['6', 'f'])]
assert table.sample == [['1', 'a'], ['2', 'b']]
assert table.read() == []
# Reset table
table.reset()
# Make assertions
assert table.read(extended=True, limit=3) == [
(2, ['id', 'name'], ['1', 'a']),
(3, ['id', 'name'], ['2', 'b']),
(4, ['id', 'name'], ['3', 'c'])]
assert table.sample == [['1', 'a'], ['2', 'b']]
assert table.read(extended=True) == [
(5, ['id', 'name'], ['4', 'd']),
(6, ['id', 'name'], ['5', 'e']),
(7, ['id', 'name'], ['6', 'f'])]
def import_resource(storage, table, schema, data):
"""Import JSONTableSchema resource to storage's table.
Parameters
----------
storage: object
Storage object.
table: str
Table name.
schema: str
Path to schema file.
data: str
Path to data file.
"""
# Create table
model = SchemaModel(schema)
schema = model.as_python
if storage.check(table):
storage.delete(table)
storage.create(table, schema)
# Write data
with topen(data, with_headers=True) as data:
storage.write(table, data)
def test_processors_chain():
# Processors
def skip_commented_rows(extended_rows):
for number, headers, row in extended_rows:
if (row and hasattr(row[0], 'startswith') and
row[0].startswith('#')):
continue
yield (number, headers, row)
def skip_blank_rows(extended_rows):
for number, headers, row in extended_rows:
if not row:
continue
yield (number, headers, row)
def cast_rows(extended_rows):
for number, headers, row in extended_rows:
crow = []
for value in row:
try:
if isinstance(value, six.string_types):
value = ast.literal_eval(value)
except Exception:
pass
crow.append(value)
yield (number, headers, crow)
# Get table
source = [['id', 'name'], ['#1', 'english'], [], ['2', '中国人']]
table = topen(source, headers='row1', post_parse=[
skip_commented_rows,
skip_blank_rows,
cast_rows])
# Make assertions
assert table.headers == ['id', 'name']
def test_html_content():
# Check raises
source = 'https://github.com/frictionlessdata/tabulator-py/blob/master/data/table.csv'
with pytest.raises(exceptions.TabulatorException) as excinfo:
table = topen(source, headers='row1')
assert 'HTML' in str(excinfo.value)
def test_web_csv_non_ascii_url():
# Get table
table = topen("http://data.defra.gov.uk/ops/government_procurement_card/over_£500_GPC_apr_2013.csv")
# Make assertions
assert table.sample[0] == ["Entity", "Transaction Posting Date", "Merchant Name", "Amount", "Description"]
def test_file_csv_parser_class():
# Get table
table = topen("data/table.csv", parser_class=CSVParser)
# Make assertions
assert table.headers is None
assert table.read() == [["id", "name"], ["1", "english"], ["2", "中国人"]]
def test_file_xls():
# Get table
table = topen('data/table.xls')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1.0, 'english'], [2.0, '中国人']]
def test_file_json_lists():
# Get table
table = topen("data/table-lists.json")
# Make assertions
assert table.headers is None
assert table.read() == [["id", "name"], [1, "english"], [2, "中国人"]]
def test_file_xls():
# Get table
table = topen('data/table.xls')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1.0, 'english'], [2.0, '中国人']]
def test_web_excel():
# Get table
table = topen(BASE_URL % 'data/table.xls')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1.0, 'english'], [2.0, '中国人']]
def test_headers_with_headers_argument():
# Get table
table = topen("data/table.csv", with_headers=True)
# Make assertions
assert table.headers == ["id", "name"]
assert list(table.iter(keyed=True)) == [{"id": "1", "name": "english"}, {"id": "2", "name": "中国人"}]
def test_file_csv_parser_options():
# Get table
table = topen('data/table.csv', parser_options={'constructor': CSVParser})
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_web_csv():
# Get table
table = topen(BASE_URL % 'data/table.csv')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_file_csv_parser_class():
# Get table
table = topen('data/table.csv', parser_class=parsers.CSV)
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_file_csv_parser_options():
# Get table
table = topen("data/table.csv", parser_options={"constructor": CSVParser})
# Make assertions
assert table.headers is None
assert table.read() == [["id", "name"], ["1", "english"], ["2", "中国人"]]
def test_file_json_lists():
# Get table
table = topen('data/table-lists.json')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1, 'english'], [2, '中国人']]
def test_file_csv_parser_class():
# Get table
table = topen('data/table.csv', parser_class=CSVParser)
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_file_xls():
# Get table
table = topen("data/table.xls")
# Make assertions
assert table.headers is None
assert table.read() == [["id", "name"], [1.0, "english"], [2.0, "中国人"]]
def test_web_json_dicts():
# Get table
table = topen(BASE_URL % 'data/table-dicts.json')
# Make assertions
assert table.headers is None
assert table.read() == [[1, 'english'], [2, '中国人']]
def test_file_json_lists():
# Get table
table = topen('data/table-lists.json')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1, 'english'], [2, '中国人']]
def test_web_excel():
# Get table
table = topen(BASE_URL % 'data/table.xls')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1.0, 'english'], [2.0, '中国人']]
def test_web_csv():
# Get table
table = topen(BASE_URL % 'data/table.csv')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_web_json_dicts():
# Get table
table = topen(BASE_URL % 'data/table-dicts.json')
# Make assertions
assert table.headers is None
assert table.read() == [[1, 'english'], [2, '中国人']]
def test_file_xls(self):
# Get table
table = topen(FPATH % 'table.xls')
# Make assertions
assert table.headers is None
assert table.read() == [('id', 'name'), (1.0, 'english'), (2.0, '中国人')]
def test_file_json_lists(self):
# Get table
table = topen(FPATH % 'table-lists.json')
# Make assertions
assert table.headers is None
assert table.read() == [('id', 'name'), (1, 'english'), (2, '中国人')]
def test_file_csv(self):
# Get table
table = topen(FPATH % 'table.csv')
# Make assertions
assert table.headers is None
assert table.read() == [('id', 'name'), ('1', 'english'), ('2', '中国人')]
def test_web_json_dicts(self):
# Get table
table = topen(WPATH % 'table-dicts.json')
# Make assertions
assert table.headers is None
assert table.read() == [(1, 'english'), (2, '中国人')]
def test_native_iterator():
# Get table
source = iter([['id', 'name'], ['1', 'english'], ['2', '中国人']])
table = topen(source)
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_text_json_lists():
# Get table
source = '[["id", "name"], [1, "english"], [2, "中国人"]]'
table = topen(source, scheme='text', format='json')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1, 'english'], [2, '中国人']]
def test_text_json_dicts():
# Get table
source = '[{"id": 1, "name": "english" }, {"id": 2, "name": "中国人" }]'
table = topen(source, scheme='text', format='json')
# Make assertions
assert table.headers is None
assert table.read() == [[1, 'english'], [2, '中国人']]
def test_text_csv():
# Get table
source = 'text://id,name\n1,english\n2,中国人\n'
table = topen(source, format='csv')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def iter(self):
'''Lazily-iterates over rows in data.
This method is useful when you don't want to load all data in memory at
once.
Returns:
iter: An iterator that yields each row in this resource.
Raises:
ValueError: If the data isn't tabular, if the resource has
no data, or if its specified encoding is incorrect
IOError: If there was some problem opening the data file (e.g. it
doesn't exist or we don't have permissions to read it).
'''
result = None
inline_data = self.metadata.get('data')
if self.local_data_path and os.path.isfile(self.local_data_path):
data_path_or_url = self.local_data_path
else:
data_path_or_url = self.remote_data_path
if inline_data:
inline_data = self._parse_inline_data()
result = iter(inline_data)
elif data_path_or_url:
dialect = self.metadata.get('dialect', {})
parser_options = {}
parser_class = None
if 'delimiter' in dialect:
parser_options['delimiter'] = dialect['delimiter']
if 'lineTerminator' in dialect:
parser_options['lineterminator'] = dialect['lineTerminator']
if len(dialect) > 0:
parser_class = tabulator.parsers.CSV
try:
table = tabulator.topen(data_path_or_url, with_headers=True,
encoding=self.metadata.get('encoding'),
parser_class=parser_class,
parser_options=parser_options)
result = TabulatorIterator(table)
except tabulator.errors.Error as e:
msg = 'Data at \'{0}\' isn\'t in a known tabular data format'
six.raise_from(ValueError(msg.format(data_path_or_url)), e)
if result is None:
if self.metadata.get('path'):
# FIXME: This is a hack to throw an IOError when local data
# exists but couldn't be loaded for some reason. If "path"
# existed and there were no issues opening it, "result" would
# never be None.
raise IOError('Resource\'s data couldn\'t be loaded.')
raise ValueError('Resource has no data')
return result
def test_native_keyed():
# Get table
source = [{'id': '1', 'name': 'english'}, {'id': '2', 'name': '中国人'}]
table = topen(source, scheme='native', format='native')
# Make assertions
assert table.headers is None
assert table.read() == [['1', 'english'], ['2', '中国人']]
def test_stream_xlsx():
# Get table
source = io.open('data/table.xlsx', mode='rb')
table = topen(source, format='xlsx')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], [1.0, 'english'], [2.0, '中国人']]
def test_stream_csv():
# Get table
source = io.open('data/table.csv', mode='rb')
table = topen(source, format='csv')
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_sample():
# Get table
source = [['id', 'name'], ['1', 'english'], ['2', '中国人']]
table = topen(source, headers='row1')
# Make assertions
assert table.headers == ['id', 'name']
assert table.sample == [['1', 'english'], ['2', '中国人']]
def test_headers_native_keyed_headers_is_none():
# Get table
source = [{'id': '1', 'name': 'english'}, {'id': '2', 'name': '中国人'}]
table = topen(source, headers=None)
# Make assertions
assert table.headers == None
assert list(table.iter(extended=True)) == [(1, None, ['1', 'english']),
(2, None, ['2', '中国人'])]
def test_save_xls(tmpdir):
# Save table
path = str(tmpdir.join('table.xls'))
table = topen('data/table.csv', headers=1)
# Assert raises
with pytest.raises(exceptions.WritingError) as excinfo:
table.save(path)
assert 'xls' in str(excinfo.value)
def test_headers_native():
# Get table
source = [[], ['id', 'name'], ['1', 'english'], ['2', '中国人']]
table = topen(source, headers='row2')
# Make assertions
assert table.headers == ['id', 'name']
assert table.read(extended=True) == [(3, ['id', 'name'], ['1', 'english']),
(4, ['id', 'name'], ['2', '中国人'])]
def test_headers_stream_context_manager():
# Get source
source = io.open('data/table.csv', mode='rb')
# Make assertions
with topen(source, headers='row1', format='csv') as table:
assert table.headers == ['id', 'name']
assert table.read(extended=True) == [(2, ['id',
'name'], ['1', 'english']),
(3, ['id', 'name'], ['2', '中国人'])]
def test_web_csv_non_ascii_url():
# Get table
table = topen(
'http://data.defra.gov.uk/ops/government_procurement_card/over_£500_GPC_apr_2013.csv'
)
# Make assertions
assert table.sample[0] == [
'Entity', 'Transaction Posting Date', 'Merchant Name', 'Amount',
'Description'
]
def test_native_generator():
# Get table
def generator():
yield ['id', 'name']
yield ['1', 'english']
yield ['2', '中国人']
table = topen(generator)
# Make assertions
assert table.headers is None
assert table.read() == [['id', 'name'], ['1', 'english'], ['2', '中国人']]
def test_processors_sample():
# Processors
def only_first_row(extended_rows):
for number, header, row in extended_rows:
if number == 1:
yield (number, header, row)
# Get table
table = topen('data/table.csv', post_parse=[only_first_row])
# Make assertions
assert table.sample == [['id', 'name']]
def test_processors_chain():
# Get table
source = [['id', 'name'], ['#1', 'english'], [], ['2', '中国人']]
table = topen(source,
headers='row1',
post_parse=[
processors.skip_commented_rows,
processors.skip_blank_rows, processors.convert_rows
])
# Make assertions
assert table.headers == ['id', 'name']
assert table.read() == [[2, '中国人']]
def test_headers_with_headers_argument():
# Get table
table = topen('data/table.csv', with_headers=True)
# Make assertions
assert table.headers == ['id', 'name']
assert list(table.iter(keyed=True)) == [{
'id': '1',
'name': 'english'
}, {
'id': '2',
'name': '中国人'
}]