Code Monkey home page Code Monkey logo

trace4cats-http4s's Issues

How to make both of`trace4cats` and `cats.effect.kernel.Ref` exist at the same time?

I wrote a simple http4s using both of trace4cats and cats.effect.kernel.Ref, but I found the type parameter challenging to give.

package example

import cats.data.Kleisli
import cats.effect.kernel.Ref
import cats.effect.{Async, ExitCode, IO, IOApp, Resource}
import cats.syntax.all._
import com.comcast.ip4s._
import fs2.Stream
import org.http4s.Http
import org.http4s.ember.server.EmberServerBuilder
import org.http4s.implicits._
import org.http4s.server.middleware.{GZip, Logger}
import org.http4s.server.{Router, Server}
import sttp.tapir.server.http4s.{Http4sServerInterpreter, Http4sServerOptions}
import trace4cats._
import trace4cats.http4s.common.Http4sRequestFilter
import trace4cats.http4s.server.syntax._
import trace4cats.jaeger.JaegerSpanCompleter


object Main extends IOApp {
  type G[x] = Kleisli[IO, Span[IO], x]

  def entryPoint[F[_]: Async](process: TraceProcess): Resource[F, EntryPoint[F]] =
    JaegerSpanCompleter[F](process = process).map { completer =>
      EntryPoint[F](SpanSampler.always, completer)
    }

  override def run(args: List[String]): IO[ExitCode] = {
    for {
      counterRef <- Ref[IO].of(0)

      serverOptions: Http4sServerOptions[G] = Http4sServerOptions.default[G]
      serverLogic: ServerLogic[G] = new ServerLogic[G](counterRef)
      resource: Resource[IO, Server] = for {
        ep <- entryPoint[IO](TraceProcess("http4s-demo"))
        r: HttpRoutes[G] = Http4sServerInterpreter[G](serverOptions).toRoutes(serverLogic.all)
        routes = r.inject(ep, requestFilter = Http4sRequestFilter.kubernetesPrometheus)
        app: Http[IO, IO] = Router("/" -> routes).orNotFound
        finalApp = Logger.httpApp(logHeaders = true, logBody = false)(GZip(app))
        s <- EmberServerBuilder.default[IO].withHost(ipv4"0.0.0.0")
          .withPort(port"8080").withHttpApp(finalApp).build >> Resource.eval(Async[IO].never)
      } yield s
      s <- Stream.resource(resource).drain.compile.drain.as(ExitCode.Success)
    } yield s
  }
}

Ref needs IO, but r needs G.

Provide call that lifts http4s client into trace context without creating new span

There is a possible minimal implementation of tracing for a http4s client, which is to lift the context to one allowing tracing, and also forwards the tracing headers; but otherwise does not create a new span or decorate the current span.

Reasons?

  • Firstly, it's a valid use case to not want the client to create a span. Maybe there are issues with forwarding trace data for this particular service, or the extra span adds unneeded extra information. However, you will still want to forward headers, so that downstream and upstream services can still be connected for the larger trace.
  • Secondly, the creation of span attributes can be something that users want to customize. E.g. if you want to match OpenTelemetry attribute names, amongst other things. By putting the monad lifting into one place, it makes adding custom behavior much simpler, as you can just use Trace[F].span etc. inside your application code. It's basically a nice building block for custom client tracing schemes.

Example code - doesn't have a Ctx parameter, but shows the general idea:

  def liftTrace[F[_]: MonadCancelThrow, G[_]: MonadCancelThrow](
      client: Client[F],
      toHeaders: ToHeaders = ToHeaders.standard
  )(implicit P: Provide[F, G, Span[F]]): Client[G] =
    Client { (request: Request[G]) =>
      Resource.eval(P.ask).flatMap { span =>
        val traceHeaders = toHeaders
          .fromContext(span.context)
          .values
          .map(Header.Raw.apply)
          .toList

        client
          .run(
            request
              .transformHeaders(_ ++ Headers(traceHeaders))
              .mapK(P.provideK(span))
          )
          .mapK(P.liftK)
          .map(_.mapK(P.liftK))
      }
    }

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.