- Moved all spark-related code into a
karadoc.spark
module. Users must replacefrom karadoc.common.run import Job
andfrom karadoc.common import Job
withfrom karadoc.spark.batch import Job
- Built-in connectors have been moved from
karadoc.connectors
tokaradoc.spark.connectors
- The default configuration parameter
model_dir
has been changed from"model"
to"models"
show_graph
command can now be called without the--tables
option. When done so, it will display the dependency graph for all models.
- Support for Python 3.6 and 3.7 has been dropped
- Removed default job import for Spark Batch: Users must replace
from karadoc.common import Job
withfrom karadoc.common.run import Job
- Added autocomplete for Karadoc, and a new command
install_autocomplete
to install it for bash