Combining multiple Case statements into single output column - sql

I need to re-categorise a column marketing_channel with 10 unique values into 15 distinct groups by matching certain criteria.
I've done this via case statements but then the output is in 15 new columns.
Is there a more elegant way to re-class the marketing_channel by simply adding 1 extra column like "marketing_sub_channel" that contains all new 15 classes?
Is there better way to do the classification than by creating 15 case statements? Was thinking a with clause, but that would also be quite lengthy
Output looks like this but ultimately just a single added column would be great:

Yes you just have to change the format a bit. Remove the "case" statement at the beginning of each line and just put the "End" at the end of the statement, like so :
CASE
WHEN condition1 THEN result1
WHEN condition2 THEN result2
WHEN conditionN THEN resultN
ELSE null
END as marketing_sub_channel
or in your case:
CASE
WHEN medium like ('%affiliate%') or marketing_cannel ='Affiliates' then 'Affiliate'
WHEN campaign like ('%_Display brand_global Progromatic Display%') then 'Dispay'
WHEN campaign like ('%display%') and campaign not like ('progrommatic') then 'Dispay'
....
else null
END as marketing_sub_channel
Also I would like to note that in your case statement since you have '%display%' and '%_Display brand_global Progromatic Display%' that you place the longer more specific one on top so it can trigger if it needs to. If '%display%' is on top then it will always trigger first since it contains a substring of the other one.

Related

SQL query find few strings in diferent columns in a table row (restrictive)

I have a table like this one (in a SQL SERVER):
field_name
field_descriptor
tag1
tag2
tag3
tag4
tag5
house
your home
home
house
null
null
null
car
first car
car
wheel
null
null
null
...
...
...
...
...
...
...
I'm developing a WIKI with a searchbar, which should be able to handle a query with more than one string for search. As an user enters a second string (spaced) the query should be able to return results that match restrictively the two strings (if exists) in any column, and so with a three string search.
Easy to do for one string with a simple SELECT with ORs.
Tried in the fronted in JS with libraries like match-sorter but it's heavy with a table with more than 100,000 results and more in the future.
I thought the query should do the heavy work, but maybe there is no simple way doing it.
Thanks in advance!
Tried to do the heavy work with all results in frontend with filtering and other libraries like match-sorter. Works but take several seconds and blocks the front.
Tried to create a simple OR/AND query but the posibilities with 3 search-strings (could be 1, 2 or 3) matching any column to any other possibility is overwhelming.
You can use STRING_SPLIT to get a separate row per search word from the search words string. Then only select rows where all search words have a match.
The query should look like this:
select *
from mytable t
where exists
(
select null
from (select value from string_split(#search, ' ')) search
having min(case when search.value in (t.tag1, t.tag2, t.tag3, t.tag4, t.tag5) then 1 else 0 end) = 1
);
Unfortunately, SQL Server seems to have a flaw (or even a bug) here and reports:
Msg 8124 Level 16 State 1 Line 8
Multiple columns are specified in an aggregated expression containing an outer reference. If an expression being aggregated contains an outer reference, then that outer reference must be the only column referenced in the expression.
Demo: https://dbfiddle.uk/kNL1PVOZ
I don't have more time at hand right now, so you may use this query as a starting point to get the final query.

Sql column value as formula in select

Can I select a column based on another column's value being listed as a formula? So I have a table, something like:
column_name formula val
one NULL 1
two NULL 2
three one + two NULL
And I want to do
SELECT
column_name,
CASE WHEN formula IS NULL
val
ELSE
(Here's where I'm confused - How do I evaluate the formula?)
END as result
FROM
table
And end up with a result set like
column_name result
one 1
two 2
three 3
You keep saying column, and column name, but you're actually talking about rows, not columns.
The problem is that you (potentially) want different formulas for each row. For example, row 4 might be (two - one) = 1 or even (three + one) = 4, where you'd have to calculate row three before you could do row 4. This means that a simple select query that parses the formulas is going to be very hard to do, and it would have to be able to handle each type of formula, and even then if the formulas reference other formulas that only makes it harder.
If you have to be able to handle functions like (two + one) * five = 15 and two + one * five = 7, then you'd be basically re-implementing a full blown eval function. You might be better to return the SQL table to another language that has eval functions built in, or you could use something like SQL Eval.net if it has to be in SQL.
Either way, though, you've still got to change "two + one" to "2 + 1" before you can do the eval with it. Because these values are in other rows, you can't see those values in the row you're looking at. To get the value for "one" you have to do something like
Select val from table where column_name = 'one'
And even then if the val is null, that means it hasn't been calculated yet, and you have to come back and try again later.
If I had to do something like this, I would create a temporary table, and load the basic table into it. Then, I'd iterate over the rows with null values, trying to replace column names with the literal values. I'd run the eval over any formulas that had no symbols anymore, setting the val for those rows. If there were still rows with no val (ie they were waiting for another row to be done first), I'd go back and iterate again. At the end, you should have a val for every row, at which point it is a simple query to get your results.
Possible solution would be like this kind....but since you mentioned very few things so this works on your above condition, not sure for anything else.
GO
SELECT
t1.column_name,
CASE WHEN t1.formula IS NULL
t1.val
ELSE
(select sum(t2.val) from table as t2 where t2.formula is not null)
END as result
FROM
table as t1
GO
If this is not working feel free to discuss it further.

How to Sum two columns meeting certain conditions

What I want to do is basically merge the two highlighted code, so that the end result is it using this SUM formula for only the items matching the LIKE criteria (under WHERE) - so that I am still able to pull GameDescriptions that do not include the LIKE criteria. Hope that makes sense... enter image description here
I think you just need to replace that part of the WHERE statement with a case statement in the SUM, like this:
SELECT
SUM(CASE WHEN GameDescription LIKE '5R25L%' THEN NetRevenue
ELSE 0 END) / COUNT(DISTINCT AccountingDate)
AS 'ES Created TheoWPU'
FROM Prime.dbo.PivotData

What is "Select -1", and how is it different from "Select 1"?

I have the following query that is part of a common table expression. I don't understand the function of the "Select -1" statement. It is obviously different than the "Select 1" that is used in "EXISTS" statements. Any ideas?
select days_old,
count(express_cd),
count(*),
case
when round(count(express_cd)*100.0/count(*),2) < 1 then '0'
else ''
end ||
cast(decimal(round(count(express_cd)*100.0/count(*),2),5,2) as varchar(7)) ||
'%'
from foo.bar
group by days_old
union all
select -1, -- Selecting the -1 here
count(express_cd),
count(*),
case
when round(count(express_cd)*100.0/count(*),2) < 1 then '0'
else ''
end ||
cast(decimal(round(count(express_cd)*100.0/count(*),2),5,2) as varchar(7)) ||
'%'
from foo.bar
where days_old between 1 and 7
It's just selecting the number "minus one" for each row returned, just like "select 1" will select the number "one" for each row returned.
There is nothing special about the "select 1" syntax uses in EXISTS statements by the way; it's just selecting some random value because EXISTS requires a record to be returned and a record needs data; the number 1 is sufficient.
Why you would do this, I have no idea.
When you have a union statement, each part of the union must contain the same columns. From what I read when I look at this, the first statement is giving you one line for each days old value and then some stats for each day old. The second part of the union is giving you a summary of all the records that are only a week or so less. Since days old column is not relevant here, they put in a fake value as a placeholder in order to do the union. OF course this is just a guess based on reading thousands of queries through the years. To be sure, I would need to actually run teh code.
Since you say this is a CTE, to really understand why this is is happening, you may need to look at the data it generates and how that data is used in the next query that uses the CTE. That might answer your question.
What you have asked is basically about a business rule unique to your company. The true answer should lie in any requirements documents for the original creation of the code. You should go look for them and read them. We can make guesses based on our own experience but only people in your company can answer the why question here.
If you can't find the documentation, then you need to talk (Yes directly talk, preferably in person) to the Stakeholders who use the data and find out what their needs were. Only do this after running the code and analyzing the results to better understand the meaning of the data returned.
Based on your query, all the records with days_old between 1 and 7 will be output as '-1', that is what select -1 does, nothing special here and there is no difference between select -1 and select 1 in exists, both will output the records as either 1 or -1, they are doing the same thing to check whether if there has any data.
Back to your query, I noticed that you have a union all and compare each four columns you select connected by union all, I am guessing your task is to get a final result with days_old not between 1 and 7 and combine the result with day_old, which is one because you take all between 1 and 7.
It is just a grouping logic there.
Your query returns aggregated
data (counts and rounds) grouped by days_old column plus one more group for data where days_old between 1 and 7.
So, -1 is just another additional group there, it cannot be 1 because days_old=1 is an another valid group.
result will be like this:
row1: days_old=1 count(*)=2 ...
row2: days_old=3 count(*)=5 ...
row3: days_old=9 count(*)=6 ...
row4: days_old=-1 count(*)=7

SQL case statement in Teradata

I'm trying to write a simple IF/ELSE statement in Teradata. As far as I understand you have to use a CASE. What I want to do is write a statement that will check if a column IS NOT NULL and display something. ELSE if it IS NULL, display something else.
Every example I found simply replaces a single value with a hard coded string or int. I'm looking for something more along the lines of this which uses the THEN statement to make another SELECT:
SELECT CARS.VIN_NUM, CARS.DRIVER_NAMES
CASE
WHEN CARS.FUEL IS NOT NULL THEN SELECT CARS.DESTINATIONS
WHEN CARS.FUEL IS NULL THEN SELECT CARS.GAS_STATIONS
END
FROM AUTOMOBILES CARS
WHERE CARS.VIN_NUM IN
('345353',
'354632',
'535231')
ORDER BY CARS.VIN_NUM
The end result should be a table displaying the VIN_NUM, DRIVER_NAMES, DESTINATIONS OR GAS_STATIONS based on the CASE. Is something like this possible or am I going about it the wrong way?
If DESTINATIONS and GAS_STATIONS are columns in the table AUTOMOBILES (aliased CARS) then the following should work:
SELECT CARS.VIN_NUM,
CARS.DRIVER_NAMES,
CASE
WHEN CARS.FUEL IS NOT NULL THEN CARS.DESTINATIONS
ELSE CARS.GAS_STATIONS
END
FROM AUTOMOBILES CARS
WHERE CARS.VIN_NUM IN
('345353',
'354632',
'535231')
ORDER BY CARS.VIN_NUM;