Hi all,
we have a data product where a data scientist built a production line and an analyst should now use the production line by setting production line variables and executing the PL.
There are cases where not all of the defined production line variables are applicable and therefore should be left empty, but this is not allowed by ONE DATA.
Is there a way to work around this? (Would a handover of the values from Apps for example allow it to have also empty spaces?)
Hello Viola,
can you provide an example scenario where it does not make sense to specify any variable value?
Maybe you can select a proper default value that has the same effect?
Kind regards
Magy
Hi Magy,
our use case is that an analyst will use the the production line built by another user and it will create a table that will be written to Oracle.
The analyst will create tables as temporary files or as permanent table. An temporary table is identified by the comment. So if the table comment is empty it is temporary, if it is filled it will be kept.
We are using production line variables for the user to insert the name of the table and also to write a comment. So for a temporary table the user would not fill the variable, but for a permanent table he would.
We do use a default value at the moment for the analyst to use, but it leads to organizational changes, because the deletion logic from Oracle will also have to be adjusted.
Best,
Viola
Hi Viola,
since a Production Line/Workflow is triggered to write the table, you could also overwrite the default value again in e.g. the Extended Mathematical Operation Processor in an extra or one of the existing Workflows, if I understood your setup correctly.
Statement would then be
CASE
WHEN comment_column = "<default_value>" THEN ""
ELSE comment_column
END
In that case you should define a default value that will most likely not be used as comment, of course. 
It will result in an empty STRING and the Oracle deletion logic could remain unchanged.
Hope that helps
Magy