Replace null value with previous available value in Row SQL server query

Question

I am looking for building a query to replace null value with previous available values.can somebody help.Here is the table currently looking like

11/30/2015          


        

Answer 1

This question is a bit old, but you can achieve the same thing using the first_row function with SQL Server (starting with 2012 version)

First, you can create a new column that contains an increasing number for each "block" of a non-null date and all the next null values:

WITH CTE AS
(
    SELECT *,
           SUM(CASE WHEN Date1 is NULL then 0 else 1 END) AS block
    FROM your_table
)

This CTE will create something like this (I'm using the column names of Shakeer's answer):

Date1          ID   Class      ID2    block
11/30/2015     ID1  ClassName   1      1
NULL           ID1  ClassName   2      1
NULL           ID1  ClassName   3      1
NULL           ID1  ClassName   4      1
12/31/2015     ID1  ClassName   5      2
NULL           ID1  ClassName   6      2
NULL           ID1  ClassName   7      2

Now, you can use the first_row function to get the first value of each "block":

SELECT *,
        first_row(Date1) OVER (PARTITION BY block ORDER BY ID2) AS NewDate
FROM CTE

I hope this helps.