Travis Crawford

Assigned to Oscar - he & Argyris are guiding the ship through treacherous cascading2 API waters.

Agreed that `ItemConverter` is not fully featured. It was enough for my needs at the time and this has never been revisited. I just added support for simple arrays in...

Not at this time. The initial use case was performing full table scans to backup tables for use in a data warehouse. What's your use case for querying GSI?

Interesting. What version of Spark are you using? Do you have a full stack trace?

What’s the issue you’re seeing trying to use this with Spark 2.2? The Spark dependency is “provided” meaning you should be able to use this jar with Spark 2.2. On...

Looking at https://github.com/traviscrawford/spark-dynamodb/blob/master/pom.xml we see there's no explicit Guava dependency, so whatever version Spark brings is what's used. Does your application override the Guava version, either explicitly or transitively from...

Hi @nadirvardar, thanks for reporting this. Can you clarify how I can reproduce? I'm happy to take a look and publish a new version.

I reproduced this locally. After the Big Merge let's see if this is still present and can take a look if so.

Is this a bug? "pants" being a launcher for the pre-build pex is the expected behavior. I'm happy to revisit this but this works the way I intended it to....

Take a look at https://rbcommons.com/s/twitter/r/95/ - that review updates commons to use a pip-installed pants rather than building from sources. When we move pants --> pantsbuild this will be the...