How to GROUP and AVG rows using LIKE

Question

I am using Crystal Reports v10. My current result set looks like this (there are 44 rows, this is an example):

item code     item desc          cost
21010DF       DOUG FIR 2x10-10   300.00
21012DF       DOUG FIR 2x10-12   310.50
21014DF       DOUG FIR 2x10-14   313.25
21016DF       DOUG FIR 2x10-16   316.10
21018DF       DOUG FIR 2x10-18   319.56
2410DF        DOUG FIR 2x4-10    271.69
2412DF        DOUG FIR 2x4-12    273.12
2414DF        DOUG FIR 2x4-14    275.98
12CDX         PLYWOOD 1/2 CDX     15.00
34TGADV       PLYWOOD T&G ADV     24.00
58CDX         PLYWOOD 5/8 CDX     18.00

I've been asked to group these items and provide an average cost for each group. The grouping rules are defined by the organization (they are random - some are logically groups while some rows are left out of of the logical group).

The result set needs to look like this:

item desc                 avg cost
DOUG FIR 2x10 (10-14)     300.00
DOUG FIR 2x10-16          316.10
DOUG FIR 2x10-18          319.56
DOUG FIR 2x4 (10-14)      271.69
PLYWOOD CDX                16.50    
PLYWOOD T&G ADV            24.00

So far, this is what I got but it doesn't seem to be coming close to what I need:

SELECT item_cd, item.item_desc, AVG(inv.cst) as avg cost
FROM item, inv
WHERE item.item_id = inv.item_id
AND item.item_cd LIKE '210%DF'
GROUP BY item.item_cd, item.item_desc

Can anyone provide advice on how to achieve this?

Answer 1

Well, if there is no logic to the grouping, then it has to be manual, sorry. In your case, for that particular result set, you can do the following:

SELECT  CASE WHEN item_cd IN ('21010DF','21012DF','21014DF') THEN 'DOUG FIR 2x10 (10-14)'
        WHEN item_cd IN ('2410DF','2412DF','2414DF') THEN 'DOUG FIR 2x4 (10-14)'
        WHEN item_cd IN ('12CDX','58CDX') THEN 'PLYWOOD CDX' ELSE item.item_desc END item_desc,
        AVG(inv.cst) as avg cost
FROM item 
LEFT JOIN inv
ON item.item_id = inv.item_id
WHERE item.item_cd LIKE '210%DF'
GROUP BY CASE WHEN item_cd IN ('21010DF','21012DF','21014DF') THEN 'DOUG FIR 2x10 (10-14)'
         WHEN item_cd IN ('2410DF','2412DF','2414DF') THEN 'DOUG FIR 2x4 (10-14)'
         WHEN item_cd IN ('12CDX','58CDX') THEN 'PLYWOOD CDX' ELSE item.item_desc END

But you should do what @jclozano said in a comment, and have the items that need to be grouped on a column marking that they belong to that specific group.