pedestal 2017-05-26 | Slack Archive

jfntn14:05:40

Is there an existing benchmark that would exercise write-body-async in the servlet-interceptor ?https://github.com/pedestal/pedestal/blob/master/service/src/io/pedestal/http/impl/servlet_interceptor.clj#L101

souenzzo14:05:19

where I find "stage" on context-map?

jfntn14:05:35

Was thinking of addressing some of the problems pointed out in https://github.com/pedestal/pedestal/issues/497 but I don’t know if there are existing benchmarks that could help measure before and after

jfntn21:05:08

We’re using the service-fn during tests but we’re still starting the server, is it possible to just bypass that?

ddeaguiar20:05:02

jfntn: Don’t call start on the created server. If you’re using component you can conditionally call start based on a setting or pass in start/stop fns during component creation. An example of the later can be found here: https://github.com/pointslope/elements/blob/master/src/pointslope/elements/pedestal.clj#L87.

ddeaguiar20:05:00

I used to be a stickler about this but I’ve become less so as of late. Now I just let the server start up using a random port.

2017-05-26

Channels