Replies: 5 comments 1 reply
-
Yeah, that's a good point. Since we've recently almost completed the database subset system and already have the reference graph defined for the tables, we can implement this feature. The feature request #153, which was resolved by MR #157, makes it possible to implement this soon. I'll aim to add this feature in version 0.2b2 by the end of August. I will create an issue with the feature request next week. |
Beta Was this translation helpful? Give feedback.
-
I've just created the feature request here #182 |
Beta Was this translation helpful? Give feedback.
-
Hi @janmeier! The major release is scheduled for October 10th! |
Beta Was this translation helpful? Give feedback.
-
@janmeier it has been implemented and merged in #229. The release will be published soon. I will keep you posted. |
Beta Was this translation helpful? Give feedback.
-
Published in v0.2.1 I would appreciate any feedback. Thanks! |
Beta Was this translation helpful? Give feedback.
-
So, we have a table named
coach
, and we want to anonymize theid
column, so that it's not possible to take an id from our staging environment and search for it in our production environment, to de-anonymize the data.However, if we do that, we would also need to apply the same transformation to all the fields that reference
coach.id
.Of course we could do that manually, but the table is referenced in many places, so it would clutter the configuration file, and we would have to remember to update the transformations every time we make a new table that references coach.
One very simple way to solve it would be to have "global" transformers defined, so that every time we encounter a field name
coachId
, it gets transformed withRandomUUID
, using thehash
engine. That would solve the issue for our usecase.However, since it looks like you are already working on analyzing foreign keys in connection with subsetting, it might make sense to do this automagically if you have all the required information available already? Maybe we could add a setting on the transformation of the
coach.id
column called something liketransformReferences: true
, and that would take care of applying the transformation to all referencing columns?Beta Was this translation helpful? Give feedback.
All reactions