You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We set defaultAssertionDataset same as defaultDataset
I am considering setting the destination for views generated by assertions to the same location as the models.
(In other words, I plan to set defaultDataset and defaultAssertionDataset to the same value.)
This is because, when defaultDataset is shared across multiple repositories, views generated by assertions may end up with duplicate or conflicting names between repositories.
Our model naming rule
For model naming, I use the following convention:
Outputs models: No prefix.
Intermediate models: Prefix _.
Source models: Prefix __.
(This is somewhat similar to Python conventions.)
I adopted this approach because prefixes like stg_ or intermediate_ might not be intuitive for non-engineer analysts who come from a marketing background.
Problem
Built-in assertions generate views with the naming convention: "${ACTION_DATASET_NAME}_${ACTION_NAME}_assertions_${BUILD_IN_ASSERTION}".
This naming pattern conflicts with the convention used for outputs models.
Solution
Introduce an additional option called buildinAssertionPrefix.
Of course, one alternative would be to create a separate defaultAssertionDataset for each repository.
However, this approach would significantly increase the number of resources, which I find less desirable.
I would appreciate it if you could kindly consider this approach.
The text was updated successfully, but these errors were encountered:
Background
We set defaultAssertionDataset same as defaultDataset
I am considering setting the destination for views generated by assertions to the same location as the models.
(In other words, I plan to set
defaultDataset
anddefaultAssertionDataset
to the same value.)This is because, when
defaultDataset
is shared across multiple repositories, views generated by assertions may end up with duplicate or conflicting names between repositories.Our model naming rule
For model naming, I use the following convention:
_
.__
.(This is somewhat similar to Python conventions.)
I adopted this approach because prefixes like
stg_
orintermediate_
might not be intuitive for non-engineer analysts who come from a marketing background.Problem
Built-in assertions generate views with the naming convention:
"${ACTION_DATASET_NAME}_${ACTION_NAME}_assertions_${BUILD_IN_ASSERTION}"
.This naming pattern conflicts with the convention used for outputs models.
Solution
Introduce an additional option called
buildinAssertionPrefix
.Of course, one alternative would be to create a separate
defaultAssertionDataset
for each repository.However, this approach would significantly increase the number of resources, which I find less desirable.
I would appreciate it if you could kindly consider this approach.
The text was updated successfully, but these errors were encountered: