SQL：插入后缺少列

Question

Basically I have two old tables, and new one which combines parts of both by matching ID as referance.

One old table is named " items_table " and contains the following columns: "Material" (which I use as ID to navigate to the items), "Condition", "Quantity", "Desc" (and a bunch of other irrelevant columns for now which I don't use)

Second old table is named " price_table " and contains: "ItemNum", "itemtype", "_subtype", "FullPrice" (and more columns which I don't want to include in my new table)

And a third new temporary table which is named " Items4sale " and should contain the following columns (values are inserted by query from the old tables): "ID", "_Type", "Subtype", "Condition", "Desc", "Quantity", "FullPrice", "Price" (this is a new calculated value), "PotentiaPrice" (this is a new calculated value)

Here's the code:

if object_id('Items4sale') is not null  /* <--- creating the new table */
drop table Items4sale;
create table Items4sale (
ID int,
_Type NVARCHAR(100),
Subtype NVARCHAR(100),
Condition INT,
Descr NVARCHAR(100),
Quantity INT,
FullPrice float, /* <--- I actually want to limit the display for prices, for example to "3299.33" instead of "3299.332120" */
Price float, 
PotentialPrice float
)

INSERT INTO Items4sale (ID, _Type, Subtype, FullPrice) /* Inserting here columns from "price_table" */
SELECT [ItemNum], [itemtype], [_subtype], [FullPrice]
FROM price_table WHERE [ItemNum] in (SELECT [Material] from items_table)  /* Same ID of the item should match to one in the items_table */
    INSERT INTO Items4sale (ID, Condition, Quantity, Descr) /* Inserting here columns from "items_table" */
SELECT [Material] (same for ID), [site_loc], [limit], [descriptions]
FROM items_table WHERE cast([site_loc] as CHAR(50)) like '%8';  /* Use only if "site_loc" column (to insert in "Condition" in the new table) ends with "8" ("5338" for example is OK and should be included) */

UPDATE Items4sale set Price=FullPrice*0.05
UPDATE items4sale set PotentialPrice=Price*Quantity

SELECT * FROM Items4sale

This is what I get as a result:

ID  |  _Type  |  Subtype | Condition  |  Descr  | Quantity | FullPrice | Price | PotentialPrice
189 |  Plastic | Toy     | NULL       | NULL    | NULL     | 8         | 0.4   | NULL
                     .........

Notice how "Condition", "Descr", "Quantity, and "PotentialPrice" come up as NULL. Those matching columns (as in the INSERT in the code above) are not null in the "items_table" where "ID" is "189".

This is the result that I wanted: (example)

 ID | _Type | Subtype | Condition | Descr | Quantity | FullPrice | Price | PotentialPrice 189 | Plastic | Toy | 3998 *(ends with '8')* | A plastic toy | 4 | 8 | 0.4 | 1.6 .........

How can I fix this?

Also, second part of my question - here's a query I ran to test for bugs, (in addition to the code above):

SELECT * FROM Items4sale WHERE ID = 189;
SELECT * FROM items_table WHERE [Material] = 189;

The above is to see if there are any missing duplicated IDs in my first query. Turns out that in "items_table" there is more than one "189" as "Material" (ID of the item) So from the second "SELECT" in the code above I get several rows with "189" as Material (ID) because there's actually more than one - I didn't take into account a column which is named "Location" (NVARCHAR(100)) in this table. I believe this is a bit more advanced: In case there's more than one row which has the ID in "items_table", then don't combine them to one row in "Items4sale" (the combined new table), but instead create them as ID = "189", "(1)189", "(2)189" (just like windows does when you try to create the same file/folder twice) yet their "Condition" (items_table.[site_loc]) in this case doesn't end with '8'. I still want to list them and apply their price and etc from "price_table".

This is important as the items might be the same (have the same barcode/ID), but have different "Location" and thus probably different "Quantity" aswell.

Answer 1

Your attempt fails because, basically, you are running 2 insert statements while you want just one record being generated.

While it should be possible to first insert with the columns from one table, then update with columns from the other one, it is simpler to perform the whole operation at once with an insert ... select query:

insert into items4sale
    (ID, _Type, Subtype, FullPrice, Condition, Quantity, Descr, Price, PotentialPrice)
select
    p.ItemNum, 
    p.ItemType, 
    p.Subtype, 
    p.FullPrice,
    i.site_loc,
    i.limit,
    i.descriptions,
    p.FullPrice * 0.05,
    p.FullPrice * 0.05 * i.limit
from items i
inner join prices p on p.ItemNum = i.Material

Unrelated note: from my understanding of your use case (you want to display derived information), this query would better fit in a view; this gives you an always-up-to-date perspective on your joined data '(on the other hand, a table requires manual updating, which can get tedious).