def test_write_timeout_error(self):
with pytest.raises(RemoteFileException):
with RemoteFile(DefaultConfig(),
"user/dataset",
"file.txt",
timeout=0.0) as writer:
writer.write("test")
def test_read_basic(self):
with responses.RequestsMock() as resp:
def download_endpoint(request):
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, "this is the test."
resp.add_callback(resp.GET, '{}/file_download/{}/{}/{}'
.format('https://query.data.world',
'user', 'dataset', 'file.txt'),
callback=download_endpoint)
with RemoteFile(DefaultConfig(), "user/dataset", "file.txt",
mode="r") as reader:
contents = reader.read()
assert "this is the test." == contents
def test_read_binary_bytes_iter(self):
with responses.RequestsMock() as resp:
def download_endpoint(request):
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, struct.pack('BBBB', 0, 1, 254, 255)
resp.add_callback(resp.GET, '{}/file_download/{}/{}/{}'
.format('https://query.data.world',
'user', 'dataset', 'file.txt'),
callback=download_endpoint)
with RemoteFile(DefaultConfig(), "user/dataset", "file.txt",
mode="rb") as reader:
contents = list(reader)
assert [b"\x00", b"\x01", b"\xfe", b"\xff"] == contents
def test_write_error(self):
with pytest.raises(RestApiError):
with responses.RequestsMock() as resp:
def upload_endpoint(request):
assert request.headers.get('User-Agent') == _user_agent()
return 400, {}, json.dumps({})
resp.add_callback(resp.PUT, '{}/uploads/{}/{}/files/{}'
.format('https://api.data.world/v0',
'user', 'dataset', 'file.txt'),
callback=upload_endpoint)
with RemoteFile(DefaultConfig(),
"user/dataset", "file.txt") as writer:
writer.write("test")
def test_read_binary_iter_chunks(self):
with responses.RequestsMock() as resp:
def download_endpoint(request):
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, "abcdef"
resp.add_callback(resp.GET, '{}/file_download/{}/{}/{}'
.format('https://query.data.world',
'user', 'dataset', 'file.txt'),
callback=download_endpoint)
with RemoteFile(DefaultConfig(), "user/dataset", "file.txt",
mode="rb", chunk_size=4) as reader:
contents = list(reader)
assert [b'abcd', b'ef'] == contents
def test_write_basic(self):
with responses.RequestsMock() as resp:
def upload_endpoint(request):
assert "test" == ''.join([chunk.decode('utf-8')
for chunk in request.body])
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, json.dumps({})
resp.add_callback(resp.PUT,
'{}/uploads/{}/{}/files/{}'
.format('https://api.data.world/v0',
'user', 'dataset', 'file.txt'),
callback=upload_endpoint)
with RemoteFile(DefaultConfig(),
"user/dataset", "file.txt") as writer:
writer.write("test")
def test_read_jsonl(self):
with responses.RequestsMock() as resp:
def download_endpoint(request):
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, '{"A":"1", "B":"2", "C":"3"}\n' \
'{"A":"4", "B":"5", "C":"6"}\n'
resp.add_callback(resp.GET, '{}/file_download/{}/{}/{}'
.format('https://query.data.world',
'user', 'dataset', 'file.csv'),
callback=download_endpoint)
with RemoteFile(DefaultConfig(), "user/dataset", "file.csv",
mode="r") as reader:
rows = [json.loads(line) for line in reader if line.strip()]
assert rows[0] == {'A': '1', 'B': '2', 'C': '3'}
assert rows[1] == {'A': '4', 'B': '5', 'C': '6'}
def test_read_csv(self):
with responses.RequestsMock() as resp:
def download_endpoint(request):
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, "A,B,C\n1,2,3\n4,5,6"
resp.add_callback(resp.GET, '{}/file_download/{}/{}/{}'
.format('https://query.data.world',
'user', 'dataset', 'file.csv'),
callback=download_endpoint)
with RemoteFile(DefaultConfig(), "user/dataset", "file.csv",
mode="r") as reader:
csvr = csv.DictReader(reader)
rows = list(csvr)
assert rows[0] == {'A': '1', 'B': '2', 'C': '3'}
assert rows[1] == {'A': '4', 'B': '5', 'C': '6'}
def test_write_csv(self):
with responses.RequestsMock() as resp:
def upload_endpoint(request):
assert "a,b\r\n42,17\r\n420,178\r\n" == \
''.join([chunk.decode('utf-8')
for chunk in request.body])
assert request.headers.get('User-Agent') == _user_agent()
return 200, {}, json.dumps({})
resp.add_callback(resp.PUT, '{}/uploads/{}/{}/files/{}'
.format('https://api.data.world/v0',
'user', 'dataset', 'file.csv'),
callback=upload_endpoint)
with RemoteFile(DefaultConfig(),
"user/dataset", "file.csv") as writer:
csvw = csv.DictWriter(writer, fieldnames=['a', 'b'])
csvw.writeheader()
csvw.writerow({'a': 42, 'b': 17})
csvw.writerow({'a': 420, 'b': 178})
def open_remote_file(self, dataset_key, file_name,
mode='w', **kwargs):
"""Open a remote file object that can be used to write to or read from
a file in a data.world dataset
:param dataset_key: Dataset identifier, in the form of owner/id
:type dataset_key: str
:param file_name: The name of the file to open
:type file_name: str
:param mode: the mode for the file - must be 'w', 'wb', 'r', or 'rb' -
indicating read/write ('r'/'w') and optionally "binary"
handling of the file data. (Default value = 'w')
:type mode: str, optional
:param chunk_size: size of chunked bytes to return when reading
streamed bytes in 'rb' mode
:type chunk_size: int, optional
:param decode_unicode: whether to decode textual responses as unicode
when returning streamed lines in 'r' mode
:type decode_unicode: bool, optional
:param **kwargs:
Examples
--------
>>> import datadotworld as dw
>>>
>>> # write a text file
>>> with dw.open_remote_file('username/test-dataset',
... 'test.txt') as w:
... w.write("this is a test.")
>>>
>>> # write a jsonlines file
>>> import json
>>> with dw.open_remote_file('username/test-dataset',
... 'test.jsonl') as w:
... json.dump({'foo':42, 'bar':"A"}, w)
... w.write("\\n")
... json.dump({'foo':13, 'bar':"B"}, w)
... w.write("\\n")
>>>
>>> # write a csv file
>>> import csv
>>> with dw.open_remote_file('username/test-dataset',
... 'test.csv') as w:
... csvw = csv.DictWriter(w, fieldnames=['foo', 'bar'])
... csvw.writeheader()
... csvw.writerow({'foo':42, 'bar':"A"})
... csvw.writerow({'foo':13, 'bar':"B"})
>>>
>>> # write a pandas dataframe as a csv file
>>> import pandas as pd
>>> df = pd.DataFrame({'foo':[1,2,3,4],'bar':['a','b','c','d']})
>>> with dw.open_remote_file('username/test-dataset',
... 'dataframe.csv') as w:
... df.to_csv(w, index=False)
>>>
>>> # write a binary file
>>> with dw.open_remote_file('username/test-dataset',
>>> 'test.txt', mode='wb') as w:
... w.write(bytes([100,97,116,97,46,119,111,114,108,100]))
>>>
>>> # read a text file
>>> with dw.open_remote_file('username/test-dataset',
... 'test.txt', mode='r') as r:
... print(r.read())
>>>
>>> # read a csv file
>>> with dw.open_remote_file('username/test-dataset',
... 'test.csv', mode='r') as r:
... csvr = csv.DictReader(r)
... for row in csvr:
... print(row['column a'], row['column b'])
>>>
>>> # read a binary file
>>> with dw.open_remote_file('username/test-dataset',
... 'test', mode='rb') as r:
... bytes = r.read()
"""
try:
return RemoteFile(self._config, dataset_key, file_name,
mode=mode, **kwargs)
except Exception as e:
raise RestApiError(cause=e)