Recipes related to leveraging Arrow Flight protocol
Suppose you want to implement a service that can store, send and receive
Parquet files using the Arrow Flight protocol,
pyarrow provides an implementation framework in :mod:`pyarrow.flight`
and particularly through the :class:`pyarrow.flight.FlightServerBase` class.
.. testcode::
import pathlib
import pyarrow as pa
import pyarrow.flight
import pyarrow.parquet
class FlightServer(pa.flight.FlightServerBase):
def __init__(self, location="grpc://0.0.0.0:8815",
repo=pathlib.Path("./datasets"), **kwargs):
super(FlightServer, self).__init__(location, **kwargs)
self._location = location
self._repo = repo
def _make_flight_info(self, dataset):
dataset_path = self._repo / dataset
schema = pa.parquet.read_schema(dataset_path)
metadata = pa.parquet.read_metadata(dataset_path)
descriptor = pa.flight.FlightDescriptor.for_path(
dataset.encode('utf-8')
)
endpoints = [pa.flight.FlightEndpoint(dataset, [self._location])]
return pyarrow.flight.FlightInfo(schema,
descriptor,
endpoints,
metadata.num_rows,
metadata.serialized_size)
def list_flights(self, context, criteria):
for dataset in self._repo.iterdir():
yield self._make_flight_info(dataset.name)
def get_flight_info(self, context, descriptor):
return self._make_flight_info(descriptor.path[0].decode('utf-8'))
def do_put(self, context, descriptor, reader, writer):
dataset = descriptor.path[0].decode('utf-8')
dataset_path = self._repo / dataset
data_table = reader.read_all()
pa.parquet.write_table(data_table, dataset_path)
def do_get(self, context, ticket):
dataset = ticket.ticket.decode('utf-8')
dataset_path = self._repo / dataset
return pa.flight.RecordBatchStream(pa.parquet.read_table(dataset_path))
def list_actions(self, context):
return [
("drop_dataset", "Delete a dataset."),
]
def do_action(self, context, action):
if action.type == "drop_dataset":
self.do_drop_dataset(action.body.to_pybytes().decode('utf-8'))
else:
raise NotImplementedError
def do_drop_dataset(self, dataset):
dataset_path = self._repo / dataset
dataset_path.unlink()
The example server exposes :meth:`pyarrow.flight.FlightServerBase.list_flights` which is the method in charge of returning the list of data streams available for fetching.
Likewise, :meth:`pyarrow.flight.FlightServerBase.get_flight_info` provides the information regarding a single specific data stream.
Then we expose :meth:`pyarrow.flight.FlightServerBase.do_get` which is in charge of actually fetching the exposed data streams and sending them to the client.
Allowing to list and dowload data streams would be pretty useless if we didn't expose a way to create them, this is the responsability of :meth:`pyarrow.flight.FlightServerBase.do_put` which is in charge of receiving new data from the client and dealing with it (in this case saving it into a parquet file)
This are the most common Arrow Flight requests, if we need to add more functionalities, we can do so using custom actions.
In the previous example a drop_dataset custom action is added.
All custom actions are executed through the
:meth:`pyarrow.flight.FlightServerBase.do_action` method, thus it's up to
the server subclass to dispatch them properly. In this case we invoke
the do_drop_dataset method when the action.type is the one we expect.
Our server can then be started with :meth:`pyarrow.flight.FlightServerBase.serve`
if __name__ == '__main__':
server = FlightServer()
server._repo.mkdir(exist_ok=True)
server.serve()
.. testcode::
:hide:
# Code block to start for real a server in background
# and wait for it to be available.
# Previous code block is just to show to user how to start it.
import threading
server = FlightServer()
server._repo.mkdir(exist_ok=True)
t = threading.Thread(target=server.serve)
t.start()
pa.flight.connect("grpc://0.0.0.0:8815").wait_for_available()
Once the server is started we can build a client to perform requests to it
.. testcode::
import pyarrow as pa
import pyarrow.flight
client = pa.flight.connect("grpc://0.0.0.0:8815")
We can create a new table and upload it so that it gets stored in a new parquet file:
.. testcode::
# Upload a new dataset
data_table = pa.table(
[["Mario", "Luigi", "Peach"]],
names=["Character"]
)
upload_descriptor = pa.flight.FlightDescriptor.for_path("uploaded.parquet")
writer, _ = client.do_put(upload_descriptor, data_table.schema)
writer.write_table(data_table)
writer.close()
Once uploaded we should be able to retrieve the metadata for our newly uploaded table:
.. testcode::
# Retrieve metadata of newly uploaded dataset
flight = client.get_flight_info(upload_descriptor)
descriptor = flight.descriptor
print("Path:", descriptor.path[0].decode('utf-8'), "Rows:", flight.total_records, "Size:", flight.total_bytes)
print("=== Schema ===")
print(flight.schema)
print("==============")
.. testoutput::
Path: uploaded.parquet Rows: 3 Size: ...
=== Schema ===
Character: string
==============
And we can fetch the content of the dataset:
.. testcode::
# Read content of the dataset
reader = client.do_get(flight.endpoints[0].ticket)
read_table = reader.read_all()
print(read_table.to_pandas().head())
.. testoutput::
Character
0 Mario
1 Luigi
2 Peach
Once we finished we can invoke our custom action to delete the dataset we newly uploaded:
.. testcode::
# Drop the newly uploaded dataset
client.do_action(pa.flight.Action("drop_dataset", "uploaded.parquet".encode('utf-8')))
.. testcode::
:hide:
# Deal with a bug in do_action, see ARROW-14255
# can be removed once 6.0.0 is released.
try:
list(client.do_action(pa.flight.Action("drop_dataset", "uploaded.parquet".encode('utf-8'))))
except:
pass
To confirm our dataset was deleted, we might list all parquet files that are currently stored by the server:
.. testcode::
# List existing datasets.
for flight in client.list_flights():
descriptor = flight.descriptor
print("Path:", descriptor.path[0].decode('utf-8'), "Rows:", flight.total_records, "Size:", flight.total_bytes)
print("=== Schema ===")
print(flight.schema)
print("==============")
print("")
.. testcode::
:hide:
# Shutdown the server
server.shutdown()
We can improve the Parquet storage service and avoid holding entire datasets in memory by streaming data. Flight readers and writers, like others in PyArrow, can be iterated through, so let's update the server from before to take advantage of this:
.. testcode::
import pathlib
import pyarrow as pa
import pyarrow.flight
import pyarrow.parquet
class FlightServer(pa.flight.FlightServerBase):
def __init__(self, location="grpc://0.0.0.0:8815",
repo=pathlib.Path("./datasets"), **kwargs):
super(FlightServer, self).__init__(location, **kwargs)
self._location = location
self._repo = repo
def _make_flight_info(self, dataset):
dataset_path = self._repo / dataset
schema = pa.parquet.read_schema(dataset_path)
metadata = pa.parquet.read_metadata(dataset_path)
descriptor = pa.flight.FlightDescriptor.for_path(
dataset.encode('utf-8')
)
endpoints = [pa.flight.FlightEndpoint(dataset, [self._location])]
return pyarrow.flight.FlightInfo(schema,
descriptor,
endpoints,
metadata.num_rows,
metadata.serialized_size)
def list_flights(self, context, criteria):
for dataset in self._repo.iterdir():
yield self._make_flight_info(dataset.name)
def get_flight_info(self, context, descriptor):
return self._make_flight_info(descriptor.path[0].decode('utf-8'))
def do_put(self, context, descriptor, reader, writer):
dataset = descriptor.path[0].decode('utf-8')
dataset_path = self._repo / dataset
# Read the uploaded data and write to Parquet incrementally
with dataset_path.open("wb") as sink:
with pa.parquet.ParquetWriter(sink, reader.schema) as writer:
for chunk in reader:
writer.write_table(pa.Table.from_batches([chunk.data]))
def do_get(self, context, ticket):
dataset = ticket.ticket.decode('utf-8')
# Stream data from a file
dataset_path = self._repo / dataset
reader = pa.parquet.ParquetFile(dataset_path)
return pa.flight.GeneratorStream(
reader.schema_arrow, reader.iter_batches())
def list_actions(self, context):
return [
("drop_dataset", "Delete a dataset."),
]
def do_action(self, context, action):
if action.type == "drop_dataset":
self.do_drop_dataset(action.body.to_pybytes().decode('utf-8'))
else:
raise NotImplementedError
def do_drop_dataset(self, dataset):
dataset_path = self._repo / dataset
dataset_path.unlink()
First, we've modified :meth:`pyarrow.flight.FlightServerBase.do_put`. Instead of reading all the uploaded data into a :class:`pyarrow.Table` before writing, we instead iterate through each batch as it comes and add it to a Parquet file.
Then, we've modified :meth:`pyarrow.flight.FlightServerBase.do_get` to stream data to the client. This uses :class:`pyarrow.flight.GeneratorStream`, which takes a schema and any iterable or iterator. Flight then iterates through and sends each record batch to the client, allowing us to handle even large Parquet files that don't fit into memory.
While GeneratorStream has the advantage that it can stream data, that means Flight must call back into Python for each record batch to send. In contrast, RecordBatchStream requires that all data is in-memory up front, but once created, all data transfer is handled purely in C++, without needing to call Python code.
Let's give the server a spin. As before, we'll start the server:
if __name__ == '__main__':
server = FlightServer()
server._repo.mkdir(exist_ok=True)
server.serve()
.. testcode::
:hide:
# Code block to start for real a server in background
# and wait for it to be available.
# Previous code block is just to show to user how to start it.
import threading
server = FlightServer()
server._repo.mkdir(exist_ok=True)
t = threading.Thread(target=server.serve)
t.start()
pa.flight.connect("grpc://0.0.0.0:8815").wait_for_available()
We create a client, and this time, we'll write batches to the writer, as if we had a stream of data instead of a table in memory:
.. testcode::
import pyarrow as pa
import pyarrow.flight
client = pa.flight.connect("grpc://0.0.0.0:8815")
# Upload a new dataset
NUM_BATCHES = 1024
ROWS_PER_BATCH = 4096
upload_descriptor = pa.flight.FlightDescriptor.for_path("streamed.parquet")
batch = pa.record_batch([
pa.array(range(ROWS_PER_BATCH)),
], names=["ints"])
writer, _ = client.do_put(upload_descriptor, batch.schema)
with writer:
for _ in range(NUM_BATCHES):
writer.write_batch(batch)
As before, we can then read it back. Again, we'll read each batch from the stream as it arrives, instead of reading them all into a table:
.. testcode::
# Read content of the dataset
flight = client.get_flight_info(upload_descriptor)
reader = client.do_get(flight.endpoints[0].ticket)
total_rows = 0
for chunk in reader:
total_rows += chunk.data.num_rows
print("Got", total_rows, "rows total, expected", NUM_BATCHES * ROWS_PER_BATCH)
.. testoutput:: Got 4194304 rows total, expected 4194304
.. testcode::
:hide:
# Shutdown the server
server.shutdown()