Is this valid "on duplicate key" syntax?

Question

I have no place to test this right now and I was hoping someone knows so I don't have to wait until tomorrow to find out....

   insert into item_properties (ItemID, PropID, Value, UpdateOn)  values 

      (538,  25, 'some description stuff goes here', unix_timestamp()),  
      (541,  25, 'some description stuff goes here', unix_timestamp()),  
      (1276, 25, 'some description stuff goes here', unix_timestamp()),  
      (1319, 25, 'some description stuff goes here', unix_timestamp()) 

   on duplicate key update  
            ItemID   = values(ItemID), 
            PropID   = values(PropID), 
            Value    = values(Value),  
            UpdateOn = values(UpdateOn)   

Can that be re-written to be:

   insert into item_properties (ItemID, Value)  values 

      (538,  'some description stuff goes here'),
      (541,  'some description stuff goes here'), 
      (1276, 'some description stuff goes here'),
      (1319, 'some description stuff goes here')

   on duplicate key update  
            ItemID   = values(ItemID), 
            Value    = values(Value),  
            PropID   = 25, 
            UpdateOn = unix_timestamp()  

Yes?

Or no, because the PropID and UpdateOn can't be accessed by the on dup part without being in the values list...?

I tried to SQLFiddle but it told me something about no DDL or DML statements, only selects.

---

So I tested the filddle...

insert into item_properties (ItemID, Value)  values 

      (538,  'some description stuff goes here'),
      (538,  'some other description stuff goes here'),
      (1276, 'some description stuff goes here'),
      (1319, 'some description stuff goes here')

   on duplicate key update  
            ItemID   = values(ItemID), 
            PropID   = 26, 
            Value    = values(Value),  
            UpdateOn = unix_timestamp()

turns into:

ITEMID       PROPID     VALUE                                    UPDATEON
538          26         some other description stuff goes here   1376952345
1276         (null)     some description stuff goes here         (null)
1319         (null)     some description stuff goes here         (null)

Which isn't the intended output...

So... I guess the two things really don't do the same thing and I need to not re-write that code in the way I initially suggested. It is valid syntax, but not correct results.

just to clarify (but I'm sure you could tell by the initial on duplicate key statement), this is the output I should end up with...

ITEMID       PROPID     VALUE                                    UPDATEON
538          26         some other description stuff goes here   1376952345
1276         26         some description stuff goes here         1376952345
1319         26         some description stuff goes here         1376952345

Thanks for the help!

Answer 1

It is valid SQL. That is to say that the above two INSERT ... ON DUPLICATE KEY UPDATE statements will have identical effects.

This can be shown via SQLFiddle, it's just that the inserts have to be part of the DDL.

http://sqlfiddle.com/#!2/56579/1/0

The ItemID = VALUES(ItemID) is also pointless if that is the duplicate key.