Devart Forums

Hi,
I am using MS SSMS v. 11.0.5058 and dbForge SQL Complete Express Edition v. 4.8.43.1. I updated dbForge recently after not having upgraded for a few years, and now notice that there is a noticable delay every time I "start working" with a table that I haven't had suggestions shown for before. Before the upgrade, the cache would be 100% there and ready instantly for suggesting. Now however, if I for example type this:

SELECT * FROM Customer where

... it will sometimes take a second for it to load the table column names on-the-fly. At worst, I will type faster than the cache loads (for example, I type "where a", because I know I want to use a column that begins with "a"), and I end up with a statement where it doesn't suggest anything useful, because the cache hadn't been loaded when I finished typing the "where" keyword.

It does this every time I begin typing with a new table name, as if it's caching the column names on-demand instead of loading the entire database's schema either upon connecting to the server or doing a "Reset Suggestions Cache" from the menu.

Am I correct that something has changed with regards to this lately, and is there a way to make the suggestions load fully?

Indeed, per lots of user requests, describing of columns is now not performed unless the user refers to the specific object in the query. This is done to speed up objects describing before the user starts typing a query in a new SQL document.

... it will sometimes take a second for it to load the table column names on-the-fly. At worst, I will type faster than the cache loads (for example, I type "where a", because I know I want to use a column that begins with "a"), and I end up with a statement where it doesn't suggest anything useful, because the cache hadn't been loaded when I finished typing the "where" keyword.

We will fix this issue in one of the next product builds.

Okay, I'm sad to hear that since that "feature" is apparently only a drawback for me. I don't understand your reply; what is gained from this change? What do you mean by "This is done to speed up objects describing"?

We just wanted to say that many users of dbForge SQL Complete complained that retrieving metadata (fetching database structure) for the code completion list takes quite a long time when the connection is slow and the databases are quite large.

Having analyzed the situation, we concluded that the largest volume of the retrieved data includes columns and parameters. Therefore, we made a deferred columns and parameters metadata retrieving. Unfortunately, there is no the corresponding option to change this back, because you are the first user who didn't like the new behavior.

Okay, it's not a huge problem thankfully.

Is there a way of downloading old versions of SQL Complete?

Previous product builds and versions are available to registered users in the Registered User's Area.

We would like to let you know that we have released a new version 5.0 of dbForge SQL Complete where the issue you reported is fixed https://www.devart.com/dbforge/sql/sqlc ... nload.html

Thank you for your help in improving dbForge SQL Complete.

Sorry for the late reply - I've been using the new version for some months now, and it seems to be pretty fast again now at suggesting. Many thanks for the fix.