PLAYP-3 #28
PLAYP-3 #28
Conversation
| final String content = context.getResponseContent(); | ||
| logger.debug("render : {}", content); | ||
| return ok(content).as(Constants.HTML_CONTENT_TYPE); | ||
| Promise<Result> promise = Promise.promise(new Function0<Result>() { |
There was a problem hiding this comment.
We have Promise<SimpleResult> promiseOfResult = Akka.future(new Callable<SimpleResult>() in the RequiresAuthenticationAction class...
Is it the same here or something different?
Definitely a change that's worth being done on 1.1.x as well...
There was a problem hiding this comment.
Yes, it's the preferred way to implement asynchronous computation. The Akka.future is deprecated in Play 2.2.0.
By the way, I was not sure it is the implementation you had in mind. it would be probably interesting that the getCredentials and getUserProfile return Futures and we compose / chain them together with map, recover... functions. But this is probably too early because we are not effectively asynchronous.
There was a problem hiding this comment.
OK. Le'ts go that way, so I think it's worth changing the way it is implemented in the RequiresAuthenticationAction at the same time...
You're right, we could have two Futures instead of just one. But I don't think it would be a great benefit (generally there is just one HTTP call for both methods). And it's still blocking HTTP calls...
I think we'll come finally to the same conclusion as for Vertx: we would need to use the non blocking HTTP client to perform these HTTP calls: https://www.playframework.com/documentation/2.0/JavaWS...
I created this JIRA: https://pac4jos.atlassian.net/browse/PACJ-9. A big change.
|
PR updated |
make callback method from CallbackController asynchronous