Andy Cutler on 30 Jun 2023 06:43:14
Hi,
It would be great to have lakehouse and warehouse support different collations, specifically case sensitive and case insensitive.
At the moment we are able to create objects that are named the same but in different casing, this can be confusing.
e.g DimProduct, dimproduct, and Dimproduct can all be created.
Kevin Conan (administrator)
We are currently working on this feature.
- Comments (6)
RE: Support Multiple Collations
Glad to hear that you are working on this feature. Wanted to throw out there - it would be a nice option to be able to configure a default collation for all objects at the tenant and/or workspace level.
RE: Support Multiple Collations
Want to second the input on being able to do case-insensitive sql for columns as well. The performance of specifying COLLATE Latin1_General_CI_AI in the where clause is awful and not as easy to pick up for some users. Thank you for working on this feature, we greatly appreciate it.
RE: Support Multiple Collations
I hope it means case-insensitive options on columns (the data in a table) and in Direct Lake semantic models.
RE: Support Multiple Collations
I see case-sensitivity in areas of Fabric as a serious mistake, particularly given the target audience.https://blog.greglow.com/2018/01/23/opinion-case-sensitivity-pox-computing/It's not just case sensitivity though. One thing that Fabric could have been is a unifying experience, yet today, each part of it has diferent rules for naming, etc. This should not have occured. It is a lost opportunity to make things easier for users.
RE: Support Multiple Collations
Case sensitivity is going to be a MAJOR obstacle to adoption in our environment. We don't have case sensitive data in all sources and trying to make everything case sensitive to migrate to fabric lake house / warehouse will be a major undertaking with no appreciable value to the end user or the business for the efforts involved.
RE: Support Multiple Collations
Case sensitivity is going to be a MAJOR obstacle to adoption in our environment. We don't have case sensitive data in all sources and trying to make everything case sensitive to migrate to fabric lakehouse / warehouse will be a major undertaking with no appreciable value to the end user or the business for the efforts involved.