How can i prevent to have same values for one user - sql

Question

I have this kind of tables:
https://ibb.co/sPn5zT7
Here in the UserPl table, the ProgrammingLanguageId and KnowledgeId are foreign keys, connected with Primary Keys of Knowledge and ProgrammingLanguage table.
I want to make when I insert for example
insert into userPLs values(1,'a7ac3486-e852-42c0-a458-9075eb5ed7d7','Doe',1,1)
here Doe says that he knows C# with basic knowledge. I want to prevent in the next insert to be impossible for Doe, to be inserted again something like this:
insert into userPLs values(1,'a7ac3486-e852-42c0-a458-9075eb5ed7d7','Doe',1,2)
or
insert into userPLs values(2,'a7ac3486-e852-42c0-a458-9075eb5ed7d7','Doe',1,2)
because he once said that his knowledge of C# is basic.
I AM USING MS SQL SERVER
How can I achieve this?

Answer 1

Try to set a unique index, where required

Answer 2

You can prevent the insert with a constraint.
alter table UserPl
add constraint UserLanguageSkillLevel
unique (UserId, ProgrammingLanguageId);
You'll still gave to catch failed inserts or modify the front end to eliminate the opportunity to add contradictory information in the first place.
A uniqueness constraint is ultimately enforced with an index. If you create a unique index directly rather than by using a constraint you could also apply the ignore_dup_key index setting and let the engine silently discard bad inserts. I'm not going to endorse that as an ideal approach but it might be useful as a temporary stopgap.

Answer 3

Having Primary key / Cluster Index on the table UserPl would enforce whatever the combination of your needs i.e.
If User cannot know multiple programming languages, then key goes
Create clustered index CLU_UserPL on UserPl (UserID)
If User can can know multiple programming languages, but cannot have multiple level of knowledge in programming languages, then key goes
Create clustered index CLU_UserPL on UserPl (UserID, ProgrammingLanguageID)
If User can can know multiple programming languages, also have multiple level of knowledge in programming languages, then key goes
Create clustered index CLU_UserPL on UserPl (RecID) --- might be new identity column
or
Create clustered index CLU_UserPL on UserPl (UserPLID)

Answer 4

this can be achieved by using constraints UNIQUE.
Here is a detailed articles about UNIQUE constraint W3School UNIQUE Article
Simple words, UNIQUE is a constraint that will ensure there is no same value allowed in the selected field.
If you want to have another way to prevent Doe to insert new values in the table, you would like to use another method which is IF EXISTS
IF EXISTS (SELECT * FROM userPLs WHERE UserId = 'THE USER ID')
BEGIN
PRINT 'Data Already Exists! Insert will be ignored!'
END
ELSE
BEGIN
PRINT 'Data doesn''t exists! Proceeding to insert the data!'
//Start inserting the data
END
UPDATED ANSWER
Here is the modified SQL Query with IF EXISTS but with another condition.
IF EXISTS (SELECT * FROM userPLs WHERE UserId = 'THE USER ID' AND ProgrammingLanguageId = 'The ID')
BEGIN
PRINT 'Data Already Exists! Insert will be ignored!'
END
ELSE
BEGIN
PRINT 'Data doesn''t exists! Proceeding to insert the data!'
//Start inserting the data
END
The query above will solve your issue. If you are wondering how does it works, below is a simple explanation:
The query will check for the UserId first. Does the UserId has been registered to Database?
Next, the query will also check, does the data that will be inserted to Database (ProgrammingLanguageId) also exists in the Database for the selected user?
If the UserId is already registered and the UserId has the same ProgrammingLanguageId with the ID that will be inserted to database, it will ignore the insert and shows "Data Already Exists! Insert will be ignored"
But IF The UserId is already registered in the Database but HAS NO ProgrammingLanguageId that match with the data that will be inserted, it will start insert the data
For a better usage, I think you should create a trigger that will occur whenever an Insert is being executed.