MySQL Collation

Hi both, thanks for the information. I have discovered that in page docs.appian.com/.../Configuring_Relational_Databases.html it says that you should use collation utf8mb4_unicode_ci but I don't know if this is for Appian data sources or for business data sources, or both?

Good find! I believe the short answer is use utf8mb4_unicode_ci - for both the primary and business databases.

Given Appian Cloud uses utf8_general_ci and the documentation says to use utf8mb4_unicode_ci, I did a quick bit of research and found this article, which suggests that utf8_general_ci shouldn't really be used anymore due to some character limitations. There's a lot more information at that link, but I'll let you read that yourself! I suppose another observation would be that Appian Cloud's MySQL configuration could be considered out-of-date; but it probably doesn't break anything... :)

Interestingly, MySQL's default collation is actually latin1_swedish_ci - it seems because the person that built MySQL was Finnish, but head of a Swedish company! In my experience utf8_general_ci is a decent enough general purpose collation, but it ultimately depends on which languages you're going to be using - or more specifically, what range of characters you expect to store.

Good find! I believe the short answer is use utf8mb4_unicode_ci - for both the primary and business databases.

Given Appian Cloud uses utf8_general_ci and the documentation says to use utf8mb4_unicode_ci, I did a quick bit of research and found this article, which suggests that utf8_general_ci shouldn't really be used anymore due to some character limitations. There's a lot more information at that link, but I'll let you read that yourself! I suppose another observation would be that Appian Cloud's MySQL configuration could be considered out-of-date; but it probably doesn't break anything... :)

Interestingly, MySQL's default collation is actually latin1_swedish_ci - it seems because the person that built MySQL was Finnish, but head of a Swedish company! In my experience utf8_general_ci is a decent enough general purpose collation, but it ultimately depends on which languages you're going to be using - or more specifically, what range of characters you expect to store.