Options: Dataset filters
CLI options for filtering the datasets in the graph
By default, the graph is filtered on the schema/dataset provided in profiles.yml. You can ignore the schema using the ignore schema option --ignore-schema=true.

Dataset filters demo

Data lineage graph generated only for upstream dependencies of 'global_mobility_report' table

Dataset filters

Sometimes the graph is too big, and you would want to filter it to more easily answer specific questions such as: - What tables are affected by a change in this table? - Where might be the source of a problem in this table? For that, you can use the following filters to get a graph that better fits your needs:
    Table filter: Filters on the selected table to see the graph that describes only this table's lineage information.
      If a partial name is defined (no schema and/or no database), then the schema and database names from the connection profile will be used.
    1
    -t, --table <TEXT>
    Copied!
    Direction: Filters on a direction of dependencies when filtering on a specific table.
      If undefined, default is both.
      This filter must be configured with a table filter.
    1
    --direction [upstream|downstream|both]
    2
    3
    # Creates the graph only for upstream lineage of 'some_schema.a_table'
    4
    edl -t some_schema.a_table --direction=upstream -d ~/dir -p my_profile
    Copied!
    Depth: Filters on how many levels of dependencies to show when filtering on a specific table.
      If undefined, default will show all dependencies.
      This filter must be configured with a table filter.
    1
    --depth <INTEGER>
    2
    3
    # Creates the graph with only 2 level depth upstream
    4
    # lineage for 'some_schema.a_table'
    5
    edl -t some_schema.a_table --depth=2 --direction=upstream -d ~/dir -p my_profile
    Copied!
Last modified 7d ago