Releases: dlt-hub/dlt
Releases · dlt-hub/dlt
0.2.0a9
0.2.0a8
What's Changed
- deletes all pipeline state (schemas, state, intermediate files) if destination dataset is dropped
- synchronizes state with the destination in the
run
method - you can opt out from state sync with
restore_from_destination=false
in ie.config.toml
- loads all schemas/source into single dataset by default. this will simplify experience for less advanced users. you can switch back to the old behavior (each source/schema has separate dataset) with
use_single_dataset=false
config option - enables CTRL-C when running user code
- commits all files extracted from several sources after all user code ran
0.2.0a7
What's Changed
- adds
max_table_nesting
argument todlt.source
to control the depth of parent-child table nesting - fixes the pipeline_name when runtime configuration is embedded
- reacts to signals (ie CTRL-C) during extraction and other signal improvements
- passes github vars to loggers/tracers
0.2.0a6
0.2.0a5
0.2.0a4
0.2.0a2
Another alpha release of 0.2.0
- is your sources and resources, specify required arguments that should come from config/secrets via
dlt.config.value
ordlt.secrets.value
- improved
init command
working with next gen of init templates - gitubpython dep fixed
0.2.0a1
This is alpha release of dlt
0.2.0. This version is a complete rewrite of the user facing interface of 0.1.0
- create sources and resources with the decorators
- credentials integration
- automatic exporting and importing of schemas
- improved syncing of schemas
- dlt state restored from the destination
and many more