I have a SQL query like this:
SELECT *
FROM (
(SELECT name FROM man)
UNION
(SELECT name FROM woman )
) AS my_table
ORDER BY name
how can I retrieve the source of my data?
For example if my result is like this:
Bob
Alice
Mario
...
I want to know if the name 'Bob' is retrieve from the 'man' table or from the 'woman' table.
SELECT *
FROM (
(SELECT name, 'man' as source FROM man)
UNION ALL
(SELECT name, 'woman' FROM woman )
) AS my_table
ORDER BY name
I added the UNION ALL becasue if these are mutually exclusive tables, it will be faster. If they are not, then adding the source will make the results mutually exclusive and you wil be able to see where the dups are. If they are not mutually exclusive but you only want to show one record, what business rule do you want to show which record you took?
A select can include a literal string, so the simplest way is probably to do:
SELECT *
FROM (
(SELECT name, 'man' as source FROM man)
UNION
(SELECT name, 'woman' as source FROM woman )
) AS my_table
ORDER BY name
These will only work if there is no intersection of Man & Woman.
If you expect duplicates, you will need to add some magic to the where clause.
and perhaps a 3rd query in the union to cover those where both exist.
Related
Hi I have a data like duplicated id but the description is different
id
name
1
A
1
B
How to remove the duplicate? since using DISTINCT will still return all the data
Its not really clear from your question in what way really you wish to remove, given that each id has different metadata attached to it.
Do you just want to de-dup the id as single column or u wish to merge its metadata together so only 1 id remains ?
The simplest one is:
select distinct id from ...
But it looks like its not you meant.
So the second option is, you merge the metadata into an array_agg:
select id, array_agg(name) as names
from (select 1 as id, 'A' as name union all select 1 as id, 'B' as name)
group by 1
This will remove the id duplication and get all metadata into an array.
If you are okay with string_agg you can go with that too:
select id, string_agg(name) as names
from (select 1 as id, 'A' as name union all select 1 as id, 'B' as name)
group by 1
This will give you comma separated values of names:
if you want fancier than this , then u can create a struct for your metadata like: (assuming you have more metadata in real project)
select id, array_agg(struct(name, info)) as metadata
from (select 1 as id, 'A' as name, 'X' as info union all select 1 as id, 'B' as name, 'Y' as info)
group by 1
this will give you:
all other options will make you lose some data, like: if you do min(name) or max(name) to consider only one row per id.
If you could clarify your question a bit better, the community can help you more. For now, I see the above options for you.
Consider below simple approach
select any_value(t).*
from your_table t
group by t.id
if you would have some extra column that identify order of entries - for example ts (timestamp) - you could use below
select any_value(t having min ts).*
from your_table t
group by t.id
select d.id, d.registration_number
from DOCUMENTS d
union
select dd.id, dd.registration_number
from DIFFERENT_DOCUMENTS dd
Would it be possible to union those results based solely on the uniqueness of the registration_number, disregarding the id of the documents?
Or, is it possible to achieve the same result in a different way?
Just to add: actually I'm unioning 5 queries, each ~20 lines long, with 4 columns that should be disregarded in determining uniqueness.
you basically need to wrap the unioned data with something else to get only the ones you want.
SELECT min(id), registration_number
FROM (SELECT id, registration_number
FROM documents
UNION ALL
SELECT id, registration_number
FROM different_documents)
GROUP BY registration_number
Union will check the combination of all the columns for uniqueness. You could, however, use union all (that does not remove duplicates) and then apply the logic yourself using the row_number window function:
SELECT id, registration_number
FROM (SELECT id, registration_number,
ROW_NUMBER() OVER (PARTITION BY registration_number ORDER BY id) AS rn
FROM (SELECT id, registration_number
FROM documents
UNION ALL
SELECT id, registration_number
FROM different_documents) u
) r
WHERE rn = 1
Since the other answers are already correct, may I ask why do you need to retrieve other columns in that query since the primary purpose appear to gather unique registration numbers?
Wouldn't it be simpler to first gather unique registration number and then retrieve the other info?
Or in your actual query, first gather the info without the columns that should be disregarded and then gather the info in these column if need be?
Like,for example, making a view with
SELECT d.registration_number
FROM DOCUMENT d
UNION
SELECT dd.registration_number
FROM DIFFERENT_DOCUMENT dd
and then gather information using that view and JOINS?
Assuming registration_number is unique in each table, you can use not exists:
select d.id, d.registration_number
from DOCUMENTS d
union all
select dd.id, dd.registration_number
from DIFFERENT_DOCUMENTS dd
where not exists (select 1
from DOCUMENTS d
where dd.registration_number = d.registration_number
);
I am trying to display the records,order as in the where clause..
example:
select name from table where name in ('Yaksha','Arun','Naveen');
It displays Arun,Naveen,Yaksha (alphabetical order)
I want display it as same order i.e 'Yaksha''Arun','Naveen'
how to display this...
I am using oracle db.
Add this ORDER BY at the query's end:
order by case name when 'Yaksha' then 1
when 'Arun' then 2
when 'Naveen' then 3
end
(There's no other way to get that order. You need an ORDER BY to get a specific result set order.)
It may be a bit clunky, but you can create a custom ordering with a case expression:
SELECT *
FROM my_table
WHERE name IN ('Yaksha', 'Arun','Naveen')
ORDER BY CASE name WHEN 'Yaksha' THEN 1
WHEN 'Arun' THEN 2
WHEN 'Naveen' THEN 3
END ASC
A slightly longer option, but one that prevents duplication of the string literals is to use a subquery:
SELECT m.*
FROM my_table m
JOIN (SELECT 'Yaksha' AS name, 1 AS name_order FROM dual
UNION ALL
SELECT 'Arun' AS name, 2 AS name_order FROM dual
UNION ALL
SELECT 'Naveen' AS name, 3 AS name_order FROM dual) o
ON o.name = m.name
ORDER BY o.name_order ASC
You can try with something like the following:
SELECT *
FROM test
WHERE name IN ( 'Yaksha', 'Arun', 'Naveen' )
ORDER BY instr ( q'['Yaksha', 'Arun', 'Naveen']', name ) ASC
This way could be useful if your IN list is somehow dynamic.
If the list of values is dynamic or you just don't want to repeat the values you could use (or abuse, depending on your point of view) a table collection, and join your real table to a table collection expression instead of using IN:
select your_table.name
from table(sys.odcivarchar2list('Yaksha','Arun','Naveen')) t
join your_table on your_table.name = t.column_value;
Which will generally work, but of course without an order-by clause is not guaranteed to work, so you can use an inline view to assign the order:
select your_table.name from (
select row_number() over (order by null) as rn, column_value as name
from table(sys.odcivarchar2list('Yaksha','Arun','Naveen'))
) t
join your_table on your_table.name = t.name
order by t.rn;
This still relies on row_number() over (order by null) using the order of the elements in the collection; which relies on collection unnesting preserving the element order. I don't think that's guaranteed either, so there is still some risk involved.
SELECT * FROM
(
SELECT city from addr_tab
UNION ALL
SELECT 'zzzz' AS city from dual
)ORDER BY city
I am using the above approach to retrieve the 'zzzz' as last record of the table. But I don't think this would work all the time, as the city theoretically might be something bigger than "zzzz" alphabetically. Is there any other robust approach to get this harcoded record as the last record?. I need this for reporting using oracle reports. Any help would be most welcome.
Add another virtual column (order_col) to use for ordering.
The query would end up something as follows.
SELECT * FROM
(
SELECT city, 1 as order_col from addr_tab
UNION ALL
SELECT 'zzzz', 2 as order_col AS city from dual
)ORDER BY order_col, city
THis way you can guarantee that 'zzzz' will always be last since the primary ordering column (order_col) is guaranteed to have a lower value for 'zzzz' than for all the rest of the records.
Well, as others have said (in comment and answer) there is probably a better way to approach the problem from the beginning; but if you need a temporary hack read on!...
Call the last record whatever you want...
'zzzz' is fine...
and then use a custom ORDER BY like:
SELECT *
FROM (
SELECT city
FROM addr_tab
UNION ALL
SELECT 'zzzz' AS city
FROM dual
)
ORDER BY
CASE WHEN city = 'zzzz' THEN 1 ELSE 0 END,
city;
this can be solve by following query also
SELECT * FROM
( SELECT city from addr_tab order by city)
UNION ALL
SELECT 'zzzz' from dual;
now got one more style to do this
with
b1 as
(
SELECT city from addr_tab order by city
)
select city from b1
union all
SELECT 'zzzz' from dual;
so any one you can use ....
Is it possible to extend query results with literals like this?
select name from users
union
select name from ('JASON');
or
select age, name from users
union
select age, name from (25,'Betty');
so it returns all the names in the table plus 'JASON', or (25,'Betty').
You use it like this:
SELECT age, name
FROM users
UNION
SELECT 25 AS age, 'Betty' AS name
Use UNION ALL to allow duplicates: if there is a 25-years old Betty among your users, the second query will not select her again with mere UNION.
In SQL Server, you would say:
Select name from users
UNION [ALL]
SELECT 'JASON'
In Oracle, you would say
Select name from user
UNION [ALL]
Select 'JASON' from DUAL
is it possible to extend query results with literals like this?
Yes.
Select Name
From Customers
UNION ALL
Select 'Jason'
Use UNION to add Jason if it isn't already in the result set.
Use UNION ALL to add Jason whether or not he's already in the result set.