Thoughts on Curd in smart platforms

In order to modify the business logic of a molecular company and modify a large number of SQL statements, it really hurts. In addition, the SQL editor of the company's existing smart platform has no function! It's just a box ~ Let alone debugging and tracking.

Go back to the smart platform project you have developed. If a business table is updated, other business tables are updated, and the value algorithms of various columns are included. What should I do at this time? Should I put curd in the background and configure it as a handwritten SQL statement, or use the background drag-and-drop Association? Let's take a look at a business scenario.

Create Financial creden

Enter basic data to the credential header, and enter the credential details

At this time, each line of details added to the UI is updated with only one action, and the detailed amount is summarized to the total amount of the creden header. The UI processing can be completed by event.

Another action: Save the credential

If it is just a simple crud, you can directly put it in the program to automatically curd. If you encounter a summary amount, you must only summarize the amount of the borrower or have more BT logic columns.

Can curd be implemented through simple control traversal in the program? Obviously, this does not work. This requires a lot of work.

I have summarized it here.

1: The Configuration Manager directly writes SQL statements for each business. For example, write SQL statements inserted into other units in unit A's SQL events.

2: Configuration Manager, full drag-and-drop configuration mode. When setting the business unit curd, other business units related to it, such as U actions, must be updated to other

Which units, specific updated fields, specific updated algorithms, pre-update and post-update?

In fact, there are still many details that need to be considered. The architecture of the smart platform is not that easy ..

Thoughts on Curd in smart platforms