Optimize Backend

[P1l3T0]

Currently, each endpoint gets its data by first getting the logged user, and then gets the needed collections. This is heavily hitting the performance of the app and leads to slow data fetching. Here is presumably the main culprit of this issue:

public async Task<User> GetUserByIdAsync(int userID)
{
    DateTime startOfMonth = new DateTime(DateTime.Now.Year, DateTime.Now.Month, 1);

    User? user = await _dataContext.Users
        .Include(u => u.Accounts)
        .Include(u => u.Categories)!
            .ThenInclude(c => c.Subcategories)
        .Include(u => u.Transactions.Where(t => t.DateCreated >= startOfMonth))
        .Include(u => u.FinancialRecommendations)
        .Include(u => u.Notifications)
        .FirstOrDefaultAsync(u => u.ID == userID);

    return user!;
}

This statement includes every data the user has (which in some cases may exceed over 100 Transactions). Refactor it so that the data is not included and it is fetched depending on the UserID:

public async Task<User?> GetUserByIdAsync(int userID)
{
    return await _dataContext.Users.FirstOrDefaultAsync(u => u.ID == userID);
}

or use a UserResponseDTO:

public async Task<CurrentUserDto?> GetUserByIdAsync(int userID)
{
    return await _dataContext.Users
        .Where(u => u.ID == userID)
        .Select(u => new CurrentUserDto
        {
            ID = u.ID,
            Name = u.Name,
            Email = u.Email,
            MethodologyType = u.MethodologyType
        })
        .FirstOrDefaultAsync();
}

or use Unit of Work Pattern

[P1l3T0]

Currently, every Controller method gets the logged User with all of their data, which could be up to 100+ entities. This results in:

• limited scalability
• bad performance over time
• lots of unneeded data being loaded

Because of this, this task needs to focus on fixing the following problems:

• rewriting core backend functionalities related to getting the User and user-specific data
• performance optimizations for the Database, mostly for larger queries using Unit of Work Pattern

As there is not much time, this will be on hold until more research is made on the above topics.