Asynchronous pure Python gRPC client and server implementation supporting asyncio, uvloop, curio and trio
Project description
purerpc
Asynchronous pure Python gRPC client and server implementation supporting asyncio, uvloop, curio and trio (achieved with anyio compatibility layer).
Requirements
- CPython >= 3.5
- PyPy >= 3.5
Installation
Latest PyPI version:
pip install purerpc
Latest development version:
pip install git+https://github.com/standy66/purerpc.git
By default purerpc uses asyncio event loop, if you want to use uvloop, curio or trio, please install them manually.
protoc plugin
purerpc adds protoc-gen-purerpc
plugin for protoc
to your PATH
enviroment variable
so you can use it to generate service definition and stubs:
protoc --purerpc_out=. --python_out=. -I. greeter.proto
or, if you installed grpcio_tools
Python package:
python -m grpc_tools.protoc --purerpc_out=. --python_out=. -I. greeter.proto
Usage
NOTE: greeter_grpc
module is generated by purerpc's protoc-gen-purerpc
plugin.
Below are the examples for Python 3.6 and above which introduced asynchronous generators as a language concept.
For Python 3.5, where native asynchronous generators are not supported, you can use async_generator library for this purpose.
Just mark yielding coroutines with @async_generator
decorator and use await yield_(value)
and await yield_from_(async_iterable)
instead of yield
.
Server
from greeter_pb2 import HelloRequest, HelloReply
from greeter_grpc import GreeterServicer
from purerpc import Server
class Greeter(GreeterServicer):
async def SayHello(self, message):
return HelloReply(message="Hello, " + message.name)
async def SayHelloToMany(self, input_messages):
async for message in input_messages:
yield HelloReply(message=f"Hello, {message.name}")
server = Server(50055)
server.add_service(Greeter().service)
server.serve(backend="asyncio") # backend can also be one of: "uvloop", "curio", "trio"
Client
import anyio
import purerpc
from greeter_pb2 import HelloRequest, HelloReply
from greeter_grpc import GreeterStub
async def gen():
for i in range(5):
yield HelloRequest(name=str(i))
async def main():
async with purerpc.insecure_channel("localhost", 50055) as channel:
stub = GreeterStub(channel)
reply = await stub.SayHello(HelloRequest(name="World"))
print(reply.message)
async for reply in stub.SayHelloToMany(gen()):
print(reply.message)
if __name__ == "__main__":
anyio.run(main, backend="asyncio") # backend can also be one of: "uvloop", "curio", "trio"
You can mix server and client code, for example make a server that requests something using purerpc from another gRPC server, etc.
More examples in misc/
folder
Release 0.5.2 (2019-07-23)
Features
- additional exception shielding for asyncio (3cbd35c)
Release 0.5.1 (2019-07-23)
Bug Fixes
- async generators on python 3.5 (1c19229)
Release 0.5.0 (2019-07-23)
Features
- can now pass contextmngr or setup_fn/teardown_fn to add_service (208dd95)
Release 0.4.1 (2019-07-22)
Features
- remove undocumented use of raw_socket in anyio (6de2c9a)
Release 0.4.0 (2019-07-22)
Bug Fixes
- speed improvements (1cb3d46)
Features
- add state property to GRPCStream (0019d8c)
- answer PING frames (c829901)
- change MAX_CONCURRENT_STREAMS from 1000 to 65536 (d2d461f)
- decouple h2 and grpclib logic (1f4e6b0)
- support percent-encoded grpc-message header (c6636f4)
- change default max message length to 32 MB
Release 0.3.2 (2019-02-15)
Bug Fixes
- fix dependencies, remove some of anyio monkey patches (ac6c5c2)
Release 0.3.1 (2019-02-15)
Bug Fixes
- fix pickling error in purerpc.test_utils._WrappedResult (9f0a63d)
Release 0.3.0 (2019-02-14)
Features
BREAKING CHANGES
- purerpc.test_utils.PureRPCTestCase is removed
Release 0.2.0 (2019-02-10)
Features
- add backend option to Server.serve (5f47f8e)
- add support for Python 3.5 (a681192)
- improved exception handling in test utils (b1df796)
- migrate to anyio (746b1c2)
BREAKING CHANGES
- Server and test now use asyncio event loop by default, this behaviour can be changed with PURERPC_BACKEND environment variable
- purerpc.Channel is removed, migrate to purerpc.insecure_channel async context manager (now supports correct shutdown)
Release 0.1.6
- Allow passing request headers to method handlers in request argument
- Allow passing custom metadata to method stub calls (in metadata optional keyword argument)
Release 0.1.5
- Enforce SO_KEEPALIVE with small timeouts
- Expose PureRPCTestCase in purerpc API for unit testing purerpc services
Release 0.1.4
- Speed up protoc plugin
Release 0.1.3 [PyPI only]
- Fix long description on PyPI
Release 0.1.2
- Fix unit tests on Python 3.7
Release 0.1.0
- Implement immediate mode
Release 0.0.1
- Initial release
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.