Copy data from backup table column to original table column - sql

Question

I accidentally updated all records for table2 data column custno with value ='33'
Now I need to recover all the column data for custno from table1 to table2 (without affecting the other data).
Note that there are no primary keys on either of the tables. It also seems several IDs are repeating.
table1 may have the same ID more than once, and all IDs need to be populated with the same date value as found in table2 for the same ID.
Any help is greatly appreciated.
-------------------so here is the actual scenario updated :-------------------------
table 1 (previous table in good state)
id col1 col2 col3 col4 colN custno
1 1 dhruv joshi 3 2 12
1 1 alpha beta 3 2 12
1 1 ebta alpha 3 2 12
1 1 dhruv joshi 3 2 11
1 1 alpha beta 3 2 11
1 1 ebta alpha 3 2 10
table 2 ( accidently updated the custno =33 for all the records )
id col1 col2 col3 col4 colN custno
1 1 dhruv joshi 3 2 33
1 1 alpha beta 3 2 33
1 1 ebta alpha 3 2 33
1 1 dhruv joshi 3 2 33
1 1 alpha beta 3 2 33
1 1 ebta alpha 3 2 33
now i have to recover this table 2 column (custno) exactly as table1 without touching any other columns.
I hope this will clear the scenario now.
After updating this query , it gives unexpected result.
UPDATE t2
SET t2.custno = t1.custno
FROM table1 AS t1
JOIN table2 AS t2
ON t1.ID = t2.ID
AND t1.col1 =t2.col1
AND t1.col2 =t2.col2
AND t1.col3 =t2.col3
AND t1.colN =t2.colN
The unexpected result being
id col1 col2 col3 col4 colN custno
1 1 dhruv joshi 3 2 12
1 1 alpha beta 3 2 12
1 1 ebta alpha 3 2 12
1 1 dhruv joshi 3 2 12
1 1 alpha beta 3 2 12
1 1 ebta alpha 3 2 12

Answer 1

In case you don't have any unique column in the table, you have to apply on condition on multiple columns (of the table) so that it makes the row uniquely identifiable for update.
Example if your table is defined as table1[ id int, col1 int, col2 varchar(100), col3 varchar(100), col4 int,..., colN int, customerNo int) and there is no unique column.
Then to update it on basis of customerNo only will pose problems.
The way is to identify multiple(or all) columns to make it more unique during update like this.
UPDATE t2
SET t2.custno = t1.custno
FROM table1 AS t1
JOIN table2 AS t2
ON t1.ID = t2.ID
AND t1.col1 =t2.col1
AND t1.col2 =t2.col2
AND t1.col3 =t2.col3
..
AND t1.colN =t2.colN
Fiddle link http://sqlfiddle.com/#!6/ff9895/2
PS: This was intended to be a comment to question but because it is too big, I am putting it as answer
I was also wondering if you had a back of the table why don't you simply restore it into original table like
-- This deletes all rows
DELETE FROM table2 WHERE Id IS NOT NULL
-- This inserts all rows from table1(the original table) into table2 to make it like table1
INSERT INTO Table2
SELECT * FROM Table1

Answer 2

I dont know much about but if tables are same structure you can try this:-
Update t2 set t2.custno = t1.custno
from
(
select ROW_NUMBER() over (order by col1 asc)as RANK, * from table_1
)t1
Inner join
(
select ROW_NUMBER() over (order by col1 asc)as RANK, * from table_2
)t2
on t1.RANK = t2.RANK