Add Columns for multiple attributes in MySQL - mysql

Question

I'm really not sure how to explain this, since I just started trying to learn how to use MySQL queries in a more advanced way today. Here is what I have:
SELECT pa.displayvalue as Brand
FROM product p
JOIN categoryheader ch ON ch.templatetype = 0 AND p.categoryid=ch.categoryid AND ch.isactive=1
JOIN categorydisplayattributes cda ON ch.headerid = cda.headerid AND ch.categoryid = cda.categoryid AND cda.templatetype = ch.templatetype AND cda.isactive=1
JOIN productattribute pa ON p.productid = pa.productid AND cda.attributeid = pa.attributeid AND pa.localeid = 1 AND pa.isactive =1
JOIN attributenames an ON pa.attributeid = an.attributeid AND an.localeid = 1
WHERE p.isactive = 1 AND p.categoryid = 4871
AND an.attributeid = 113319;
So I have that query, and literally, all I want to do is add another column for the "And an.attributeid = 113319"
I just want to add another column but instead of 113319, I want it to pull the values of let's say 1762 into a column next to it.
So column 1 would have all the values that "And an.attributeid = 113319" pulls, and column two would have all the values that "And an.attributeid = 1762" pulls.
I got this code to work kind of:
SELECT pa.displayvalue as Brand
FROM product p
JOIN categoryheader ch ON ch.templatetype = 0 AND p.categoryid=ch.categoryid AND ch.isactive=1
JOIN categorydisplayattributes cda ON ch.headerid = cda.headerid AND ch.categoryid = cda.categoryid AND cda.templatetype = ch.templatetype AND cda.isactive=1
JOIN productattribute pa ON p.productid = pa.productid AND cda.attributeid = pa.attributeid AND pa.localeid = 1 AND pa.isactive =1
JOIN attributenames an ON pa.attributeid = an.attributeid AND an.localeid = 1
WHERE p.isactive = 1 AND p.categoryid = 4871
AND an.attributeid = 113319;
SELECT pa.displayvalue as Type
FROM product p
JOIN categoryheader ch ON ch.templatetype = 0 AND p.categoryid=ch.categoryid AND ch.isactive=1
JOIN categorydisplayattributes cda ON ch.headerid = cda.headerid AND ch.categoryid = cda.categoryid AND cda.templatetype = ch.templatetype AND cda.isactive=1
JOIN productattribute pa ON p.productid = pa.productid AND cda.attributeid = pa.attributeid AND pa.localeid = 1 AND pa.isactive =1
JOIN attributenames an ON pa.attributeid = an.attributeid AND an.localeid = 1
WHERE p.isactive = 1 AND p.categoryid = 4871
AND an.attributeid = 1762;
But this is returned in navicat as two seperate results. I want both of these to be in one result but as two columns, instead of two results with one column each.
As you can see, all of the lines are the same in each of those Selects except that last line.
I also feel like there is a better way than to have to use that whole code twice just to get another column for an.attributeid. Regardless, I'll take what I can get.
If you have ANY suggestions, I would appreciate it very very much.

Answer 1

If you really want to do that, for formatting or reporting purpose. This is a hack you would use.
And the use of LEFT JOIN would return all rows from the left table, with the matching rows in the right table. The result is NULL in the right side when there is no match.
SELECT p.productid, t2.Type, t1.Brand, t3.Resolution
FROM product p
JOIN
(
SELECT p.productid, pa.displayvalue as Brand
FROM product p
JOIN categoryheader ch ON ch.templatetype = 0 AND p.categoryid=ch.categoryid AND ch.isactive=1
JOIN categorydisplayattributes cda ON ch.headerid = cda.headerid AND ch.categoryid = cda.categoryid AND cda.templatetype = ch.templatetype AND cda.isactive=1
JOIN productattribute pa ON p.productid = pa.productid AND cda.attributeid = pa.attributeid AND pa.localeid = 1 AND pa.isactive =1
JOIN attributenames an ON pa.attributeid = an.attributeid AND an.localeid = 1
WHERE p.isactive = 1 AND p.categoryid = 4871
AND an.attributeid = 113319
) t1 ON t1.productid = p.productid
LEFT OUTER JOIN
(
SELECT p.productid, pa.displayvalue as Type
FROM product p
JOIN categoryheader ch ON ch.templatetype = 0 AND p.categoryid=ch.categoryid AND ch.isactive=1
JOIN categorydisplayattributes cda ON ch.headerid = cda.headerid AND ch.categoryid = cda.categoryid AND cda.templatetype = ch.templatetype AND cda.isactive=1
JOIN productattribute pa ON p.productid = pa.productid AND cda.attributeid = pa.attributeid AND pa.localeid = 1 AND pa.isactive =1
JOIN attributenames an ON pa.attributeid = an.attributeid AND an.localeid = 1
WHERE p.isactive = 1 AND p.categoryid = 4871
AND an.attributeid = 1100
) t2 ON t2.productid = p.productid
LEFT OUTER JOIN
(
SELECT p.productid, pa.displayvalue as Resolution
FROM product p
JOIN categoryheader ch ON ch.templatetype = 0 AND p.categoryid=ch.categoryid AND ch.isactive=1
JOIN categorydisplayattributes cda ON ch.headerid = cda.headerid AND ch.categoryid = cda.categoryid AND cda.templatetype = ch.templatetype AND cda.isactive=1
JOIN productattribute pa ON p.productid = pa.productid AND cda.attributeid = pa.attributeid AND pa.localeid = 1 AND pa.isactive =1
JOIN attributenames an ON pa.attributeid = an.attributeid AND an.localeid = 1
WHERE p.isactive = 1 AND p.categoryid = 4871
AND an.attributeid = 1762
) t3 ON t3.productid = p.productid

Answer 2

You can do something like this
... AND an.attributeid IN (113319, 1762 )
This would return you two rows with one column.
If you want to transpose rows into columns, then that gets a bit more complicated.

Answer 3

This is sometimes called a key-value store. You need to play some games with multiple joins to do what you want.
This sort of thing will do the trick
SELECT whatever,
attr1.displayvalue attr1,
attr2.displayvalue attr2
FROM product p
LEFT JOIN productattribute attr1 ON p.productid = attr1.productid
AND attr1.attributeid = 113319
AND attr1.localeid = 1
AND attr1.isactive = 1
LEFT JOIN productattribute attr2 ON p.productid = attr2.productid
AND attr2.attributeid = 1762
AND attr2.localeid = 1
AND attr2.isactive = 1
You end up joining two copies of the attribute table, and selecting the appropriate attribute id from each in the join condition. You use left join rather than ordinary join so you still get the row even if the attribute is missing.
A very common example in the wild of this data organization is the wp_postmeta table in WordPress.