Ahh I didn't know this operator ! Thank you very much :)
I did a test and it works (it's much less code):
def detect(data: ByteString) = {
// ...Apache Tika stuff...
Some(MediaType.custom("text/plain", binary =
true).asInstanceOf[MediaType.Binary])
}
def detect(source: Source[ByteString, _], previousPrefix: ByteString =
ByteString.empty): Future[(MediaType.Binary, Source[ByteString, _])] =
source.prefixAndTail(1).runWith(Sink.head).flatMap { case (prefix, tail) =>
detect(prefix.head).fold {
detect(tail, prefix.head)
}{ mediaType =>
Future.successful((mediaType, tail.prepend(Source.single(previousPrefix
++ prefix.head))))
}
}
[...]
val route =
path("test") {
get {
complete("123456789\n")
}
} ~
(path("media") & parameter("url" ? "http://127.0.0.1:8080/test";)) { url =>
get {
complete(Http().singleRequest(HttpRequest(uri = url)).flatMap(response =>
detect(response.entity.dataBytes)).map { case (mediaType, source) =>
HttpEntity(ContentType(mediaType), source)
})
}
}
With `prefixAndTail` I ask for only 1 `ByteString` and try to detect the
mime-type with it, but if it's not enough, the `detect` method is executed
until I can figure out the mime-type.
I did some benchmarks with `wrk`, here is the results:
$ wrk -t8 -c32 -d10s "http://127.0.0.1:8080/test";
Running 10s test @ http://127.0.0.1:8080/test
8 threads and 32 connections
Thread Stats Avg Stdev Max +/- Stdev
Latency 4.58ms 23.80ms 437.09ms 97.11%
Req/Sec 4.89k 1.21k 11.45k 75.73%
386406 requests in 10.04s, 56.38MB read
Requests/sec: 38499.53
Transfer/sec: 5.62MB
$ wrk -t8 -c32 -d10s "http://127.0.0.1:8080/media";
Running 10s test @ http://127.0.0.1:8080/media
8 threads and 32 connections
Thread Stats Avg Stdev Max +/- Stdev
Latency 15.24ms 26.91ms 338.60ms 92.50%
Req/Sec 449.52 86.50 820.00 71.46%
35837 requests in 10.04s, 5.33MB read
Requests/sec: 3571.02
Transfer/sec: 544.06KB
It's a naïve benchmark but it's just to give me an idea about the
implementation.
Thank you again :)
Victor
Le mercredi 5 avril 2017 23:59:03 UTC+2, Kyrylo Stokoz a écrit :
>
> I'm not sure but maybe Source.prefixAndTail can help you to archive same
> thing?
>
> You can run detection on prefix and later combine it with tail again via
> source.concat?
>
--
>>>>>>>>>> Read the docs: http://akka.io/docs/
>>>>>>>>>> Check the FAQ:
>>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html
>>>>>>>>>> Search the archives: https://groups.google.com/group/akka-user
---
You received this message because you are subscribed to the Google Groups "Akka
User List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.
