snowplow / snowplow-scala-analytics-sdk

Scala SDK for working with Snowplow enriched events in Spark, AWS Lambda, Flink et al.
https://snowplow.github.io/snowplow-scala-analytics-sdk/
20 stars 14 forks source link

SBT StackOverflowError #77

Closed NielsAlebregtse closed 5 years ago

NielsAlebregtse commented 5 years ago

We are trying to incorporate the snowplow-scala-analytics-sdk into one of our projects. However, we keep running into a StackOverflowError when we try to compile our source code with sbt. Trying a simple sbt clean test on a clone of the current sdk master repo yields the same error on my machine.

Can you give us some guidance how to use the new version of the snowplow-scala-analytics-sdk?

chuwy commented 5 years ago

Hi @NielsAlebregtse, that's very strange. What versions of JDK and SBT you're using?

bekiroguz commented 5 years ago

Writing about the same problem: We use JDK8 and sbt 1.2.8. Apparently sbt compile started working again after passing -Xss2M parameter in SBT_OPTS env variable, or adding -J-Xss2M line to the sbtopts file located in /usr/local/etc/sbtopts in my macbook.

bekiroguz commented 5 years ago

This should be related to the circe library macro's trying to process the Event case class with 131 fields.

chuwy commented 5 years ago

Thanks @bekiroguz. Yes, codec derivation is an expensive operation. I was just surprised because it worked on multiple machines we tried before, without any issues. Closing this.