SQL "IN" clause usage to include / exclude items at the same time? [closed] - sql

Question

Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 4 years ago.
Improve this question
I would like to have some help on the best way to address a specific revision. I have the following SQL query:
SELECT * FROM PRODDTA.F4105
WHERE (COITM, COMCU) IN ((`"Set of Data 1"`))
AND (COLOCN, COLEDG) NOT IN ((`"Set of Data 2"`));
What I want to do is to revise if there is someway to merge these items in one:
SELECT * FROM PRODDTA.F4105
WHERE (COITM, COMCU, COLOCN, COLEDG) IN ((`"Set of Data 1 & 2"`));
Do you know if this is achievable?
Thanks a lot in advance,

Answer 1

This really depends on where Set of data 1 and Set of data2 come from and what their relationship is to each other, and I'm pressed to imagine a real world scenario where this would be a good idea.
A Set of Data here is really just a set of Records with Fields. These Fields have, inherently, a relationship to one another which is why they are present on a single Record.
Issue 1: The two sets contain data that don't relate to each other:
Set of Data1 could be completely disparate from Set of Data2 and so there is no way to say that they can be combined.
Imagine two sets:
Set1:
+-------+--------+
| color | animal |
+-------+--------+
| brown | dog |
| white | dog |
| black | dog |
| green | parrot |
| green | turtle |
| brown | turtle |
+-------+--------+
Set2:
+------------+--------+
| food | flavor |
+------------+--------+
| pepper | spicy |
| water | |
| grapefruit | bitter |
| lemon | sour |
| candy | sweet |
+------------+--------+
A queries WHERE clause:
WHERE (f1, f2) IN (SELECT color, animal FROM set1)
AND (f3, f4) IN (SELECT food, flavor FROM set2)
There's no good way to write this where we test (f1, f2, f3, f4) as there is no relationship between color | animal and food | flavor.
We could, if we were in crazytown, cross join the two sets to get their cartesian product yielding the same result set as the original query:
WHERE (f1, f2, f3, f4) IN (SELECT color, animal, food, flavor FROM set1, set2)
But now we have a subquery with an intermediate result set of set1 x set2 number of records. This is dumb for multiple reasons:
Indexes on the two sets are ignored
If set1 or set2 are more than a few records, you end up with a HUGE intermediate result set.
There is no relationship between these two sets, so it's crazy town to combine them just to make your SQL less characters.
There will be huge amounts of unnecessary system resources (CPU, Disk, I/O) used to build and store this intermediate result set resulting in a cumbersome slow query.
Any other developer upon seeing it will hunt you down and kill you. If they call me up, I'll provide the getaway car.
Issue 2: The two sets could have a relationship with one another, but changing your IN conditions into 1 causes records to drop.
Even if the two sets can be combined, you may still end up with results that differ from the original query. Imagine:
Set1:
+-------+--------+
| color | animal |
+-------+--------+
| brown | dog |
| white | dog |
| black | dog |
| green | parrot |
| green | turtle |
| brown | turtle |
+-------+--------+
Set 2:
+--------+-------------+
| animal | stink_scale |
+--------+-------------+
| turtle | 2 |
| parrot | 4 |
| dog | 5 |
| skunk | 10 |
+--------+-------------+
Table1:
+-------+--------+---------+-------------+
| color | animal | animal2 | stink_scale |
+-------+--------+---------+-------------+
| white | dog | dog | 5 |
| brown | dog | parrot | 4 |
| green | turtle | turtle | 2 |
+-------+--------+---------+-------------+
The query you want to change:
SELECT * FROM table1
WHERE (color, animal) IN (SELECT color, animal FROM Set1)
AND (animal2, stink_scale) IN (SELECT animal, stink_scale FROM set2);
This would yield 3 records as white | dog is in set1 and dog | 5 is in set as it brown | dog in set1 and parrot | 4 is in set2 and the same for the third record in table1.
BUT if we combine these two sets on their animal key:
SELECT set1.color, set1.animal, set2.animal as animal2, set2.stink_scale FROM set1 JOIN set2 ON set1.animal = set2.animal;
We will get the set:
+-------+--------+---------+-------------+
| color | animal | animal2 | stink_scale |
+-------+--------+---------+-------------+
| brown | dog | dog | 5 |
| white | dog | dog | 5 |
| black | dog | dog | 5 |
| green | parrot | parrot | 4 |
| green | turtle | turtle | 2 |
| brown | turtle | turtle | 2 |
+-------+--------+---------+-------------+
And we use that to combine our IN conditions:
SELECT *
FROM table1
WHERE (color, animal, animal2, stink_scale) IN (SELECT set1.color, set1.animal, set2.animal as animal2, set2.stink_scale FROM set1 JOIN set2 ON set1.animal = set2.animal)
We only get 2 records back since there is no result in that subquery where brown | dog | parrot | 4 will exist.
So, in the end unless there is a reason to change the conditions, thus changing the definition of the result set, you probably best off leaving it alone. It really changes the logic.