At the moment, we have in the blockchain parser model classes a lot of Pojos based on Hadoop Writable. This issue is about separating them in dedicated classes. This avoids Hadoop dependencies for Big Data frameworks, such as Apache Flink, that remove Hadoop as a required dependency.
jornfranke
commented
3 years ago
At the moment, we have in the blockchain parser model classes a lot of Pojos based on Hadoop Writable. This issue is about separating them in dedicated classes. This avoids Hadoop dependencies for Big Data frameworks, such as Apache Flink, that remove Hadoop as a required dependency.