如何根据其他列的值将列的值增加 1 否则不会增加该值？

Question

I have a table with column (difference).

difference  tableNumber

     10    
     22    
     03    
     45    
     32    
     120   
     65    
     34    
     85    
     90    
     130   
     140   
     65    
     21    
     01    
     45    
     150   
     55    
     5000  
     87    
     43    

The first rows tableNumber will start with '1'.

difference  tableNumber

     10          1
     22    
     03    
     45    
     32    
     120   
     65    
     34    
     85    
     90    
     130   
     140   
     65    
     21    
     01    
     45    
     150   
     55    
     5000  
     87    
     43    

Now, (from second row onwards) if the 'difference' < 100 then insert the 'tableNumber' of previous rows' value

else if the 'difference' > 100 then increment the value of 'tableNumber' of previous row by 1 and insert it in the current row.

Edit: Sample table: (tableNumber is the expected output)

So, this will be the expected output:

 difference tableNumber 

     10          1         
     22          1      
     03          1      
     45          1      
     32          1      
     120         2      
     65          2      
     34          2      
     85          2      
     90          2      
     130         3      
     140         4      
     65          4      
     21          4      
     01          4      
     45          4      
     150         5      
     55          5      
     5000        6      
     87          6      
     43          6      

Answer 1

SQL tables represent unordered sets. What you want to do seems to depend on the ordering of the rows. I will assume that you have a column that specifies the ordering. Let's call it ord .

I think the following basically does what you want:

select difference,
       1 + sum(case when difference < 100 then 0 else 1 end) over (order by ord) as tablenumber
from t;

This does not take into account the first "1" in the table. However, that seems superfluous to the problem you really want to solve (identifying adjacent groups where the values are less than 100).

If that is important, you can replace the 1 + with something like (select difference from t order by ord fetch first row only) + -- or whatever expression is appropriate for your database.