async-rpc
=========
Non-blocking XML-RPC client for Python. Provides base classes
``async_rpc.BaseSerializer`` and ``async_rpc.BaseServerProxy``,
you can write your own implementation of RPC (JSON RPC, …).
Usage
-----
.. code-block:: python
from async_rpc import XmlRpcServerProxy
async def get_data():
proxy = XmlRpcServerProxy('http://example.com/RPC2:8000')
response = await proxy.methodName(123, 456)
return response
API documentation
-----------------
*class* async_rpc.\ **BaseRpcServerProxy**\ (*uri, timeout=None,
max_clients=None, user_agent=None, use_dns_cache=None, ttl_dns_cache=None,
http_version=None, keepalive_timeout=None, \*\*kwargs*)
Base class for server proxies. It is responsible for HTTP transport. For
concrete RPC implementation uses descendant of the *BaseSerializer*.
- **uri** (*str*) – URL of the remote server
- **timeout** (*float*) – timeout in seconds for network operations,
includes DNS requests, default is 1.0 seconds.
- **max_clients** (*int*) – size of the connections pool,
default is 16 connections.
- **user_agent** (*str*) – user agent which is sent to server,
default is ``"Python async-rpc"``.
- **use_dns_cache** (*bool*) – use internal cache for DNS lookups,
default is ``True``.
- **ttl_dns_cache** (*float*) – store DNS lookups into internal cache
for a *ttl_dns_cache* seconds, default is 10.0 seconds.
- **http_version** (*str*) – version of the HTTP protocol, can be
``"1.0"`` or ``"1.1"``, default is ``"1.0"``.
- **keepalive_timeout** (*float*) – close connection after *keepalive_timeout*
seconds, if ``None``, keep-alive is disabled, default is ``None``.
- **kwargs** (*dict*) – additional keyword arguments, which will be passed
into serializer constructor.
*attribute* async_rpc.BaseRpcServerProxy.\ **serializer_cls**
(*BaseSerializer*) – serializer class
*coroutine* async_rpc.BaseRpcServerProxy.\ **call**\ (*name, \*\*params*)
- **name** (*str*) – mame of the remote function.
- **params** (*dict*) – parameters, which will be passed to remote function.
Instance of the **BaseServerProxy** is callable, so you can call remote
function directly on server proxy instance. These calls are equivalent:
.. code-block:: python
res = await proxy.getData('12345678901234567890')
res = await proxy.call('getData', '12345678901234567890')
*class* async_rpc.\ **BaseSerializer**\ (*\*\*kwargs*)
Ancestor for concrete implementation of RPC. Contains four abstract methods,
which must be overriden in inherited classes.
- *kwargs* (*dict*) – additional keyword arguments, which are passed from
server proxy.
*method* async_rpc.BaseSerializer.\ **initialize**\ (*\*\*kwargs*)
Initialize instance. It is called from constructor when instance is
created.
- **kwargs** (*dict*) – additional keyword arguments, which are passed from
server proxy.
*abstractmethod* async_rpc.BaseSerializer.\ **prepare_request_headers**\ (
*params*)
Return ``dict`` containig HTTP headers. Method is called before RPC call
is sent. You can add additional HTTP header, which are be sent in request
to remote server.
- **params** (*tuple*) – params for RPC call.
*abstractmethod* async_rpc.BaseSerializer.\ **dumps**\ (*response_headers*)
Return data, which will be sent as POST body in request to remote
server. Method is called before RPC call is sent.
- **params** (*tuple*) – params for RPC call.
- **menthodname** (*str*) – name of the RPC method.
*abstractmethod* async_rpc.BaseSerializer.\ **process_response_headers**\ (
*response_headers*)
Check headers from response. Method is called after response is
received from server.
- **response_headers** (*Mapping*) – mapping containing response
HTTP headers.
*abstractmethod* async_rpc.BaseSerializer.\ **loads**\ (*data,
response_headers*)
Return response from remote server as Python objects. Method is
called after response is received from server.
- **data** (*bytes*) – response body.
- **response_headers** (*Mapping*) – mapping containing response
HTTP headers.
License
-------
3-clause BSD