Using Flask I can define something like
@self.app.before_request
def before_request():
if 'Authorization' not in request.headers:
abort(401)
if request.method == 'POST':
if not request.json:
abort(400)I was wondering how I can do it with nameko
Hey @antoniodimariano. The exact equivalent of this would be to create a custom entrypoint that inherited from the built-in HTTP one, and modified the handle_request logic (see https://github.com/nameko/nameko/blob/master/nameko/web/handlers.py#L40)
However, you may want to be more granular and check these things for certain entrypoints only. For that I would use a decorator around the service method that inspected the
request object before maybe calling the wrapped method. There's an example of that here: https://github.com/nameko/nameko-amqp-retry/blob/master/nameko_amqp_retry/decorators.py#L30
I moved to nameko after a few years with Node.js
so, I would rather say I am still in the 'learning mode'
but happy to know that we are in the same boat
so AFAIK nameko is a good solution, but I ‘ve not testing it in production yet
I'm not really using Nameko as my API server, mainly using it for running smaller services for computations etc. Still using NodeJS as my primary API server. Although I do have a few APIs being served from Nameko which are suddenly giving me issues (Although the errors might be because I am missing something)
if you wanna share your piece of code,
is there a way to dynamically define a RpcProxy of a service ? Because I don't know which service will depeend on anotyher one in advance. I made a project like IFTTT.com where you can plug service with each other but of course, you don't know what service will depend on another one
or show I just use nameko.event instead of rpc ?
how dinamically would you like to define a rpc?
as soon as you plug in a new service you define a new proxy to use?
ffs, I would like to have more documentation and more examples
real use example I mean
not the common stuff
...
for exampe, anyone know how I can disable the debug message in nameko
I am using the logging module
but I have the console full of msgs like
2017-12-19 15:35:46,472 nameko.containers DEBUG signalling result for <WorkerContext [proxy.register_action] at 0x105ce79b0>
tore down worker <WorkerContext [proxy.register_action] at 0x105ce79b0> in 0.000s
in order to dinamically assign a proxy?
@antoniodimariano, add a logging configuration to your config.yaml and set the level of the root logger there. example: http://nameko.readthedocs.io/en/stable/cli.html?highlight=logging#running-a-service
@foxmask there is a recent thread on the mailing list exactly about dynamic RPC proxies: https://groups.google.com/forum/?utm_medium=email&utm_source=footer#!topic/nameko-dev/XE-uarCEV00
runner.add_service() i meant
@foxmask re containers, yes -- the container is the thing that runs a singular service. it manages the extensions (entrypoints, dependency providers) declared by the service and creates workers when an entrypoint fires. the runner is just a convenience wrapper around one or more containers.
when you do
nameko run <module> on the command line, nameko will discover any service classes defined in <module> and start them all under one runner
I was wondering where I can find a nice guide to logging