Say I have a table showing the type of fruit consumed by an individual over a 24 hour period that looks like this:
Name Fruit
Tim Apple
Tim Orange
Tim Orange
Tim Orange
Lisa Peach
Lisa Apple
Lisa Peach
Eric Plum
Eric Orange
Eric Plum
How would I get a table that shows only the most consumed fruit for each person, as well as the number of fruits consumed. In other words, a table that looks like this:
Name Fruit Number
Tim Orange 3
Lisa Peach 2
Eric Plum 2
I tried
SELECT Name, Fruit, Count(Fruit)
FROM table
GROUP BY Name
But that returns an error because Name needs to be in the GROUP BY statement as well. Every other method I've tried returns the counts for ALL values rather than just the maximum values. MAX(COUNT()) doesn't appear to be a valid statement, so I'm not sure what else to do.
This is a pain, but you can do it. Start with your query and then use join:
SELECT n.Name, n.Fruit
FROM (SELECT Name, Fruit, Count(Fruit) as cnt
FROM table as t
GROUP BY Name, Fruit
) as t INNER JOIN
(SELECT Name, max(cnt) as maxcnt
FROM (SELECT Name, Fruit, Count(Fruit) as cnt
FROM table
GROUP BY Name, Fruit
) as t
GROUP BY Name
) as n
ON t.name = n.name and t.cnt = n.maxcnt;
Related
Not sure how to explain this..
I have a similar table, but i have simplified it with the following:
I have a table of goods shipped to different cusotmers. Some have bought apples only, others have bought apples and potates.
I want an SQL query to return only customers where "To be billed" = Yes AND the customer hasnt bought any vegetables.
So for example if the table looks like this:
Item
Name
Group
To_be_billed
CustomerNo.
2000
Apple
Fruit
Yes
1
2000
Apple
Fruit
No
2
2000
Apple
Fruit
No
3
2000
Apple
Fruit
Yes
4
2000
Apple
Fruit
Yes
5
4000
Potato
Vegetable
No
2
4000
Potato
Vegetable
No
4
I want the query to return:
Item
Name
Group
To_be_billed
CustomerNo.
2000
Apple
Fruit
Yes
1
2000
Apple
Fruit
Yes
5
The reason 4 has bought apples, and is to be billed, but the customer also bought Potatoes, so is to be ignored...
You can create a CTE to check for CustomerNo.s that you need to ignore, and then use not exists:
with bought_veg as
(
select "CustomerNo."
from tbl
where tbl."Group" like 'Vegetable'
)
select tbl.*
from tbl
where not exists (select 1 from bought_veg where tbl."CustomerNo." = bought_veg."CustomerNo.")
and tbl.To_be_billed = 'Yes'
Example without CTE:
select tbl.*
from tbl
where not exists (select "CustomerNo." from tbl t2 where tbl.[CustomerNo.] = t2.[CustomerNo.] and "Group" like 'Vegetable')
and tbl.To_be_billed = 'Yes'
I have a Postgres table with the following format:
Fruit
Owner
Apple
John
Orange
Susan
Pear
Michael
Peach
Susan
I want to write a query that creates a new column, Owner's Fruits, which lists all fruits owned by the owner of the row in question, like this:
Fruit
Owner
Owner's Fruits
Apple
John
Apple
Orange
Susan
Orange, Peach
Pear
Michael
Pear
Peach
Susan
Orange, Peach
However, I am unable to figure out how to do this, nor even how to phrase my question so that I find an answer on here or Google.
It seems like it should be simple enough. How can I do this?
You can use string_agg():
select t.*,
string_agg(fruit, ',') over (partition by owner) as owners_fruits
from t;
That said, I actually recommend putting the values into an array instead:
select t.*,
array_agg(fruit) over (partition by owner) as owners_fruits
from t;
Arrays are more versatile.
Here is my table. I am trying to get distinct person distinct fruit based on maximum number of fruits he has.
persons | fruits
David apple
David apple
David apple
David banana
David orange
Sam apple
Sam banana
Sam orange
Sam orange
Sam orange
Sam orange
Tom apple
Tom banana
Tom banana
Tom orange
I want to see my result as:
persons | fruits
David apple
Sam orange
Tom banana
I tried using count and max functions and group by, but was not able to get right result.
You can use distinct on:
select distinct on (person) person, fruit
from (select person, fruit, count(*) as cnt
from personfruits pf
group by person, fruit
) pf
order by person, cnt desc;
You can write this without the subquery as well:
select distinct on (person) person, fruit
from personfruits pf
group by person, fruit
order by person, count(*) desc;
However, that is a bit hard to follow for someone not really familiar with distinct on.
From what I understand, you want to see which fruit occurs most often, per person. If that's correct, this should work
SELECT persons, fruits
FROM (
SELECT
persons,
fruits,
RANK() OVER(PARTITION BY persons ORDER BY FruitCount DESC) AS FruitRank -- Rank fruit count per person
FROM (
SELECT
persons,
fruits,
count(*) FruitCount -- get # rows per (person, fruit) combination
FROM MyTable
GROUP BY persons, fruits
) src
) src
WHERE FruitRank = 1 -- Return fruit with largest FruitCount, per person
If I have a structure like this:
CREATE TABLE things (
id,
personA varchar2,
personB varchar2,
attribute ...,
)
And I want to find, for a given attribute, if I have at least 1 common person for all my things, how would I go about it?
So if my data is (and it could be more than 2 per attribute):
1, John, Steve, Apple
2, Steve, Larry, Apple
3, Paul, Larry, Orange
4, Paul, Larry, Orange
5, Chris, Michael, Tomato
6, Steve, Larry, Tomato
For Apple, Steve is my common person, For Orange both Paul and Larry are, and for Tomato I have no common people. I don't need a query that returns all of these at once, however. I have one of these attributes and want 0, 1, or 2 rows depending on what kind of commonality I have. I've been trying to come up with something but can't quite figure out.
This will give you your common person / attribute list. I ran it against your sample data and got the expected result. Hope it's at least pointing in the right direction :)
WITH NormNames AS (
SELECT PersonA AS Person, Attribute FROM things
UNION ALL SELECT PersonB AS Person, Attribute FROM things
)
SELECT
Person, Attribute, COUNT(*)
FROM NormNames
GROUP BY Person, Attribute
HAVING COUNT(*) >= 2
If you're on 11gR2 you could also use the unpivot operator to avoid the self-join:
select person, attribute
from (
select *
from things
unpivot (person for which_person in (persona as 'A', personb as 'B'))
)
group by person, attribute
having count(*) > 1;
PERSON ATTRIBUTE
---------- ----------
Steve Apple
Paul Orange
Larry Orange
3 rows selected.
Or to just the the people who match the attribute, which I think is what the end of your question is looking for:
select person
from (
select *
from things
unpivot (person for which_person in (persona as 'A', personb as 'B'))
)
where attribute = 'Apple'
group by person, attribute
having count(*) > 1;
PERSON
----------
Steve
1 row selected.
The unpivot translates columns into rows. Run on its own it transforms your original six rows into twelve, replacing the original persona/personb columns with a single person and an additional column indicating which column the new row was formed from, which we don't really care about here:
select *
from things
unpivot (person for which_person in (persona as 'A', personb as 'B'));
ID ATTRIBUTE W PERSON
---------- ---------- - ----------
1 Apple A John
1 Apple B Steve
2 Apple A Steve
2 Apple B Larry
3 Orange A Paul
3 Orange B Larry
4 Orange A Paul
4 Orange B Larry
5 Tomato A Chris
5 Tomato B Michael
6 Tomato A Steve
6 Tomato B Larry
12 rows selected.
The outer query is then doing a simple group.
Here's one method.
It implements an unpivot method by cross-joining to a list of numbers (you could use the unpivot method Alex uses) and then joins the result set, hopefully with a hash join for added goodness.
with
row_generator as (
select 1 counter from dual union all
select 2 counter from dual),
data_generator as (
select
attribute,
id ,
case counter
when 1 then persona
when 2 then personb
end person
from
things,
row_generator)
select
t1.attribute,
t1.person
from
row_generator t1,
row_generator t2
where
t1.attribute = t2.attribute and
t1.person = t2.person and
t1.id != t2.id;
I've searched for an answer on this but can't find quite how to get this distinct recordset based on a condition. I have a table with the following sample data:
Type Color Location Supplier
---- ----- -------- --------
Apple Green New York ABC
Apple Green New York XYZ
Apple Green Los Angeles ABC
Apple Red Chicago ABC
Apple Red Chicago XYZ
Apple Red Chicago DEF
Banana Yellow Miami ABC
Banana Yellow Miami DEF
Banana Yellow Miami XYZ
Banana Yellow Atlanta ABC
I'd like to create a query that shows the count of unique locations for each distinct Type+Color where the number of unique locations is more than 1, e.g.
Type Color UniqueLocations
---- ----- --------
Apple Green 2
Banana Yellow 2
Note that {Apple, Red, 1} doesn't appear because there is only 1 location for red apples (Chicago). I think I've got this one (but perhaps there is a simpler method). I'm using:
SELECT Type, Color, Count(Location) FROM
(SELECT DISTINCT Type, Color, Location FROM MyTable)
GROUP BY Type, Color HAVING Count(Location)>1;
How can I create another query that lists the Type, Color, and Location for each distinct Type,Color when the count of unique locations for that Type,Color is greater than 1? The resulting recordset would look like:
Type Color Location
---- ----- --------
Apple Green New York
Apple Green Los Angeles
Banana Yellow Miami
Banana Yellow Atlanta
Note that Apple, Red, Chicago doesn't appear because there is only 1 location for red apples. Thanks!
Use a COUNT(DISTINCT Location) and join against a subquery on Type and Color The GROUP BY and HAVING clauses as you have attempted to use them will do the job.
/* Be sure to use DISTINCT in the outer query to de-dup */
SELECT DISTINCT
MyTable.Type,
MyTable.Color,
Location
FROM
MyTable
INNER JOIN (
/* Joined subquery returns type,color pairs having COUNT(DISTINCT Location) > 1 */
SELECT
Type,
Color,
/* Don't actually need to select this value - it could just be in the HAVING */
COUNT(DISTINCT Location) AS UniqueLocations
FROM
MyTable
GROUP BY Type, Color
/* Note: Some RDBMS won't allow the alias here and you
would have to use the expanded form
HAVING COUNT(DISTINCT Location) > 1
*/
HAVING UniqueLocations > 1
/* JOIN back against the main table on Type, Color */
) subq ON MyTable.Type = subq.Type AND MyTable.Color = subq.Color
Here is a demonstration
You could write your first query as this:
Select Type, Color, Count(Distinct Location) As UniqueLocations
From Table
Group By Type, Color
Having Count(Distinct Location) > 1
(if you're using MySQL you could use the alias UniqueLocations in your having clause, but on many other systems the aliases are not yet available as the having clause is evaluated before the select clause, in this case you have to repeat the count on both clauses).
And for the second one, there are many different ways to write that, this could be one:
Select Distinct Type, Color, Location
From Table
Where
Exists (
Select
*
From
Table Table_1
Where
Table_1.Type = Table.Type
and Table_1.Color = Table.Color
Group By
Type, Color
Having
Count(Distinct Location) > 1
)