If you are in a middle-ware that both receives the context and maybe append some data to context to send it to the next interceptor, then which of the two methods i.e. metadata.FromOutgoingContext and metadata.FromIncomingContext shall be called?
If you are writing that middle-ware in the server, then you are receiving that metadata in the incoming request.
You should then use metadata.FromIncomingContext to get the metadata at that point.
The metadata in the "outgoing context" is the one generated by the client when sending an outgoing request to the server.
See here for examples of both:
https://github.com/grpc/grpc-go/blob/master/Documentation/grpc-metadata.md
Related
Looking to make a service call based on a header value.
I can see two options:
1) Do it from the controller which is mainly used for a different service.
2) Add a filter which will do this by reading the request context.
Want to know what's the best way to handle this in a Spring application.
I guess it depends on your requirements.
Controller:
If you have to make service calls for preparing the response of a specific controller. For example, you have a controller say:
/employee
fand or preparing the response of this endpoint you need to call say staff service.
In this case, it's better to handle such calls in controllers.
Filter:
If you want to intercept each request and perform some operation on the request before sending it to the controller or before sending the response to the client.
A use case here can be checking the roles of the user by intercepting all requests.
As we know by using the filter, we can perform two operations at two instances −
Before sending the request to the controller
Before sending a response to the client.
In this case, OncePerRequestFilter is quite useful from the spring web module.
Quoting the documentation :
Filter base class that aims to guarantee a single execution per request dispatch, on any servlet container.
I am using spring boot 2 and apache camel 2.24 to build an api gateway which exposes REST end points to receive JSON/XML request and do the following
Validate incoming JSON/XML
Convert incoming request to downstream expected format.
Push the request to a camel route which invokes the downstream REST end point and returns back the response.
Convert response to expected format and return back to client.
Currently I have route configured as below
from("direct::camel").process(preprocessor).process(httpClientProcessor).process(postprocessor);
httpClientProcessor - This does the main job of invoking downstream end point and return the response.
preprocessor - splits the request for configured requests, puts them in a list before passing to httpClientProcessor.
postprocessor - does the following based on content type
XML - Remove "xml" tag from individual responses and combine to form one single response under one root element
JSON - Combine json response under one parent array element.
There can be cases where the multiple requests need to be sent to the same end point or each to be sent to a unique end point. Currently I have that logic in httpClientProcessor. One issue with this approach is that I can invoke the downstream end points only one after another rather than in parallel (unless I add thread pool executor in httpClientProcessor)
I am new to apache camel and hence started with this basis route configuration. Based on going through the documentation, I came across camel components like split(), parallelProcessing(), multicast and aggregator components of camel but I am not sure how to plug these all together to achieve my requirement -
Split the incoming request using a pre configured delimiter to create multiple requests. The incoming request may or may not have multiple requests.
Based on endpoint url configuration, either post all split requests to same end point or each request to a unique endpoint.
Combine responses from all to form one master response (XML or JSON) as output from the route.
Please advise.
This sounds like you should have a look a the Camel Split EIP.
Especially
How to split a request into parts and re-aggregate the parts later
How to process each part in parallel
For the dynamic downstream endpoints, you can use a dynamic To (.toD(...)) or the Recipient List EIP. The latter can even send a message to multiple or no endpoint.
I am creating an asynchronous application using zeromq, celluloid. I need to maintain the states for different tasks that depend on some response. I can do it by sending data about the state in the response params. But is there any better way to do this?
Use uniquely identified Conditions to wait for replies.
The question is very abstract/vague, so I will do the best I can to give you a functional example. If I understand/guess correctly, this is what you need to do:
Generate a UUID with Celluloid::Internals::UUID.generate before the request is made.
Package the request data, and send the UUID with the data.
Create a Condition, associated with that UUID.
Send the request.
Process the request, keeping the UUID associated with it.
Return the response, sending the UUID back with the reply data.
Process the response on a generic ( non-stated ) level... then signal the Condition with the data returned, locating the Condition in a thread-safe Hash by UUID.
Receive the data you needed within the stated object, by blocking at wait in that stated object... rather than processing the response directly.
The stated object ought to have no awareness of 0MQ which is by design stateless... it ought to communicate with a layer which makes requests, receives responses, and delivers the response to the object who asked for it, without that stated object losing its state, and without the state information being passed in the request or the response.
This strategy is used heavily by ECell which is forthcoming. Hopefully you are already using the celluloid-zmq gem for evented 0MQ support already.
I want to send a message to a Queue using the JMS Component. However, I don't want to send the payload message, but just a simple String with the server date time, using the MEL expression #[server.dateTime].
How can I accomplish that?
If you put your JMS Component inside an Async Component, you will get a clone of the original payload. From this point you can do anything you want.
Mule Enricher is another way to get this work done without disturbing the current payload..
Just put set payload with server.dateTime and send it to JMS outbound inside an Enricher .. it will work without disturbing or loosing the actual payload
Reference :- https://developer.mulesoft.com/docs/display/current/Message+Enricher
and
Mule Enrichment: enrich xml payload with http endpoint response
Pls note : if you use multiple message processor inside an Enricher, you need to put inside <processor-chain/>
I am attempting to implement a firefox extension which filters incoming HTTP responses and handles them accordingly.
I wish to block HTTP responses containing a specific header. Having read through some of the MDC articles, it appears the most common way of observing HTTP responses is by registering an http-on-examine-response observer.
My question is: Using this observer, is it possible to block the HTTP request from further processing (as a by-product, the page will continue to wait for a response)? Or do I need to use some other facet of XPCOM (if so, care to point me in the right direction)?
Thanks in advance,Mike
Should be possible: the "subject" of that notification is a HTTP channel, on which you can call nsIRequest::cancel().
the page will not continue to wait for a response, though, since you blocked the response. Perhaps you've been actually looking for a way to alter an HTTP response?