Mixing 3 queries

[Canman2005]

Hi all

 

I have 3 QUERIES, they are

 

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 0 GROUP BY `product_id`

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 3 GROUP BY `type_uid`

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 4 GROUP BY `type_uid`

 

Is it possible to combine all these 3 QUERIES into one?

 

I have tired UNION and JOIN but cant seem to get it to work.

 

Any help would be ace

 

Thanks

 

Dave

[Mchl]

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 0 GROUP BY `product_id`
UNION
SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 3 GROUP BY `type_uid`
UNION
SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 4 GROUP BY `type_uid`

??

 

 

What exactly you want results to be?

[DarkerAngel]

Could I have more information on what exactly your trying to accomplish?

 

I see that your trying to have the three queries combined into one but is there a specific reason behind this combination.

 

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND `type_type` = 0 OR `type_type` = 3 OR `type_type` = 4 GROUP BY `type_uid`

Just something off the top of my head...

 

You can have multiple comparisons for the same column.

[Mchl]

DarkAngel: Your query will never return any results.

[DarkerAngel]

lol yeah... your correct... I meant to put 'OR'

[Mchl]

lol yeah... your correct... I meant to put 'OR'

 

Then you should probably add some parentheses as well

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND (`type_type` = 0 OR `type_type` = 3 OR `type_type` = 4) GROUP BY `type_uid`

 

But as long as Canman2005 doesn't tell us what he wants, we can't be sure

[Canman2005]

Basically I have a log of products purchase, if value `type_type` equals 0 then it should return all rows but doing a GROUP BY `product_id` so that I only return products that do not have the same `product_id`

 

If the value of `type_type` is equal to 3 or 4, then it should return all rows but do a group GROUP BY `type_uid` as that means those rows with the same `type_uid` are a group of products

 

Does that make sense?

 

Thanks

[DarkerAngel]

Ok, but with what you said it would only make sense to logically combine the query if the "type_type" = 3 or 4, and leave it at the one query if it equals 0

[Canman2005]

So is there no possible way to do it? Would I need to do 3 seperate QUERIES and not combine the results

[DarkerAngel]

here lets try this, give me your `log` structure and just some random sample data, I don't want to steal your code lol I have to much of my own to worry about, I just want to get a better feel of what your trying to accomplish and show you better queries.

 

But:

SELECT * FROM `log` WHERE `unique_id` = '51006401' AND (`type_type` = 3 OR `type_type` = 4) GROUP BY `type_uid`

 

Should work to combine the GROUP BY `type_uid` queries, I still don't quite understand what your trying to do... lol but I'm trying to help.

[Canman2005]

Basically I have a log table which contains `product information`

 

If a row contains the value 3 or 4 under the field `type_type` then it's classed as a group of products, so more than one product

 

If a row contains the value 0 under the field `type_type` then it's classed as a single product

 

Okay, so if any rows have 0 then they should be listed as normal but doing a GROUP BY `product_id` as I want to group together all single products with the same `product_id` value, basically so I can do a 'quanity' value if more than 1 of the same `product_id` is stored.

 

If a row has 3 or 4 under `type_type` then I want to group the results together with GROUP BY `type_uid`, the reason is that you might have 10 products with the same `type_uid` code, that means they are a group of products and should be grouped together.

 

Does that help anymore?

[Canman2005]

So is the only method to do 3 seperate QUERIES?

[Mchl]

Two actually. One for type_type being 3 or 4  and another for type_type being 0

[fenway]

Ugh... you can't use * with GROUP BY...still.