Currently Cloudberry supports AsterixDB as its backend using a translator to SQL++. To make Cloudberry more general, we want to support other backends as well.
In https://github.com/ISG-ICS/cloudberry/issues/287 we finished the step of supporting SparkSQL, and its core is really SQL. In this issue, we want to fully support SQL on top of the previous PR. Specifically:
Step 1: Support write operators using SQL;
Step 2: Setup a TwitterMap using MySQL as the backend.
[x] summarize the differences between the SparkSQL with the MySQL queries. Depend on the number of differences, we will decide the design strategy. The SQL query strings are located in the SparkSQLGeneratorTest
Currently Cloudberry supports AsterixDB as its backend using a translator to SQL++. To make Cloudberry more general, we want to support other backends as well.
In https://github.com/ISG-ICS/cloudberry/issues/287 we finished the step of supporting SparkSQL, and its core is really SQL. In this issue, we want to fully support SQL on top of the previous PR. Specifically:
Step 1: Support write operators using SQL; Step 2: Setup a TwitterMap using MySQL as the backend.