We also should improve SQL DataProvider files for Turkish_CI_AS collation.

Turkish people reporting that DNN can be installed succesfully under this server & db collation, but when it comes to playing, it fails at some points.

I will try to get more info about this issue.

This is very helpful. *I* am not a Core Trustee so I am not able to do anything to move this forward but you have posted this in the right place to be considered.

However, be prepared for a possible delay in implementation because the time consuming task is evaluating what this change has on the rest of the project, meaning will it break something else? Also the other enhancements and fixes that might be affected have to be considered. For example a rewrite of core functionality might make this section scheduled to be replaced in 6 months. In that case it would not be worth it to spends hours testing this particular change (I don't think this is the case I'm just giving an example).

So it's not just as simple as dropping it in. This happens to the code of other Core members. It's good code but doesn't make it in.

But is may be used so thank you for taking the time to contribute to the project.

I intend going through the "code for review" suggestions early next year, so I'll evaluate these further at the time. Please note, we usually have to take time to assess the suggestions to ensure that they do not have negative effects e.g. regarding using stringbuilder for concatenations, it's true that it's more efficent for large concatenation operations, but there is an overhead with creating the stringbuilder object itself - i had to assess this for a past project and it turned out that any concatenation task with less than 10-12 concatenations performs worse with the stringbuilder as opposed to using "&".

Cathal