90% of the time I want to see what documents were created most recently. I have a "date_created" attribute that I'd like to be able to sort on by default.

Clicking the header in the collection only sorts the documents shown, not all the documents in the collection.

The new dark mode looks and feels quite bad. Can you please see how Sequel Pro (a free product for MySQL) developed it in comparison to Studio 3T?

I would give 8/10 to Sequel Pro and 2/10 to Studio 3T for the execution of dark mode.

BTW: I am on Mac, me and my team love Studio 3T, but team laughed at the initial release...

If I start writing a query manually in the QBE fields (Query, Projection, Sort, Skip, Limit), I lose those entries if I open Visual Query Builder and start to add query parameters.

Would be awesome if I could switch freely between QBE, Visual Query Builder, and an editable Query Code view, with my query preserved across the three input methods.

If I want a task to run more than once per day, I have to clone a separate instance of the task for each scheduled run. This is particularly tedious if I want something to run every 10 minutes, 24/7/365.

Would be MUCH better if I had full cron-like control over the task schedules.

Tasks and Scheduler are very handy for prototyping. Would be even better if S3T could generate scripts for executing the tasks in test or production environments, and without requiring S3T to be running on a desktop at the time the scheduled task is to run. (Maybe a command-line S3T, or S3T task-runtime package?)

Possibility to set breakpoints and debug more complex database scripts in the script editor.

If you run a script twice in intelli shell, and you use a cons x = 3, you cannot run the script again.

[js] TypeError: redeclaration of const x :

even if you change it to a var afterwards, it still produces that error.
It seems that the context is kept between running scripts?

I suggest that at the very least the document number for which the error occurred be provided. Line number or position within document would be very helpful.

Indenting blocks of code at once would be useful.

Special whitespace characters that sneak in via clipboard result in phantom errors. Having a toggle to show markers in place of all whitespace characters would be good. Lots of editors have this feature.

Maybe swapping your editor with a complete third-party (e.g. as in vscode) would be a good idea - if that sounds doable (less maintenance on your part, and you stay up to date in trends).

There is no find and replace in the script editor.

Situation: You are looking at a small collection of very small JSON documents so you select JSON View. You then need to open a new collection. However, this is a is giant collection with huge JSON documents (multiple arrays of subdocuments). It's so big it can take up to 30 seconds to render and or crash Studio3T altogether. But you can't stop the operation. So just have to wait. What would fix this is an option for default collection view when opening a new collection. I would just set that to Tree View and this situation would happen anymore.

Please, parse "Expire After" field in months, days, hours, minutes and seconds when user being set "Expire After" field.

I have multiple Mongo instances on different hosts that cannot, for security reasons, connect to each other to enable schema compare. Ideally, I'd like to save a "typical" schema description (collection names, indexes, etc) to a file that I could then use to compare against actual Mongo instances.

It would be good to be able to synchronise any changes from the schema / file automatically.

Currently when I drill down through a document into an array field, clicking the headers implies that the table is now sorted by that element, but in fact it looks to be sorted by the entire array field (basically array[0] is sorted instead).

When I'm running a long running operation sometimes I have to go home, so I have to abort and start over in the morning. It would be nice to pause, and continue later.

Can you please add an option to fully normalize a mongodb collection into several SQL tables without PK value repetition?

meaning for example movies collection:

title, country
James Bond 19 [USA,UK,Spain]
Avatar2 [USA]

into:

movies:
id, title
0 James Bond 19
1 Avatar 2

countries:
id country
0 USA
1 UK
2 Spain

moviecounries
movieid country_id
0 0
0 1
0 , 2
1 , 0

• currently migration will create two "USA" values with different country id which is redundant.

Currently the SQL migration feature populates a default "_id" PK.
I would like to be able to choose my own PK from the migrated columns.

Would love to see a visual data modeling tool similar to Hackolade with both forward and reverse-engineering capabilities.