Here is my problem. I have a Microsoft SQL server database table with a bunch of columns in it. (I cannot change the data structure of this table!)
21 of the columns in this table stand for 'Actual' sizes
21 of the columns in this table stand for 'Relative' sizes
The Actual size columns correspond through Relative size columns via column number. (ex : ActualColumn1 corresponds to RelativeColumn1, ActualColumn2 corresponds to RelativeColumn2, ActualColumnX to RelativeColumnX up to 21)
Based on the inputted 'Actual' value I need to return the related 'Relative' value.
I have managed to do part of what I need with the following case statement, but case statements have a maximum of 10 conditions and therefore won't be able to cover all 21 columns in my table. What would be the best way to approach this issue? My goal would be to put this into a select statement so that I could select the Relative Size and return it in my query.
Example of what I did with the case statement:
SELECT
T.AValue
CASE WHEN (T.ActualColumn1 = T.AValue) then T.RelativeColumn1 ELSE
CASE WHEN (T.ActualColumn2 = T.AValue) THEN T.RelativeColumn2 ELSE
CASE WHEN (T.ActualColumn3 = T.AValue) THEN T.RelativeColumn3 ELSE
CASE WHEN (T.ActualColumn4 = T.AValue) THEN T.RelativeColumn4 ELSE
NULL
END
END
END
END AS RValue
FROM T
Thank you for your help!
If you change your case statement to a searched case you can have more then 10 when clauses.
select T.AValue,
case T.AValue
when T.ActualColumn1 then T.RelativeColumn1
when T.ActualColumn2 then T.RelativeColumn2
when T.ActualColumn3 then T.RelativeColumn3
when T.ActualColumn4 then T.RelativeColumn4
end as RValue
from T
Related
I am trying to replace a range of values with a string. I know how to do it with the replace function but that, as far as I know, requires them to be done one at a time.
Is there a way to select a range of values, for example (1-200), and replace them with a singular string value say "BLANK"?
I have tried WHEN, THEN and SET but get a syntax error near WHEN or SET as I try these.
Base Code Idea
Select DATA
WHEN DATA >= 1 THEN 'BLANK'
WHEN DATA <200 THEN 'BLANK
END
FROM DATABANK
Thanks!
Is this what you want?
select data,
case when data not between 1 and 200 then data end as new_data
from databank
What this does is take the integer value of data, and replace any value that's in the 1-200 range with null values, while leaving other values unchanged. The result goes into column new_data.
The assumption here is that data is a number - so the alternative value has to be consistent with that datatype (string 'BLANK' isn't): I went for null, which is consistent with any datatype, and is the default value returned by a case expression when no branch matches.
If you wanted something else, say 0, you would do:
select data,
case when data between 1 and 200 then 0 else data end as new_data
from databank
I need to select one and only 1 row of data based on an ID in the data I have. I thought I had solved this (For details, see my original question and my solution, here: PostgreSQL - Select only 1 row for each ID)
However, I now still get multiple values in some cases. If there is only "N/A" and 1 other value, then no problem.. but if I have multiple values like: "N/A", "value1" and "value2" for example, then my case statement is not sufficient and I get both "value1" and "value2" returned to me. This is the case statement in question:
CASE
WHEN "PQ"."Value" = 'N/A' THEN 1
ELSE 0
END
I need to give a unique integer value to each string value and then the problem will be solved. The question is: how do I do this? My first thought is to somehow convert the character values to ASCII and sum them up.. but I am not sure how to do that and also worried about performance. Is there a way to very simply assign a value to each string so that I can choose 1 value only? I don't care which one actually... just that it's only 1.
EDIT
I am now trying to create a function to add up the ASCII values of each character so I can essentially change my case statement to something like this:
CASE
WHEN "PQ"."Value" = 'N/A' THEN 9999999
ELSE SumASCII("PQ"."Value")
END
Having a small problem with it though.. I have added it as a separate question, here: PostgreSQL - ERROR: query has no destination for result data
EDIT 2
Thanks to #Bohemian, I now have a working solution, which is as follows:
CASE
WHEN "PQ"."Value" = 'N/A' THEN -1
ELSE ('x'||LPAD(MD5("PQ"."Value"),16,'0'))::bit(64)::bigint
END DESC
This will produce a "unique" number for each value:
('x'||substr(md5("PQ"."Value"),1,8))::bit(64)::bigint
Strictly speaking, there is a chance of a collision, but it's very remote.
If the result is "too big", you could try modulus:
<above-calculation> % 10000
Although collisions would then be a 0.01% chance, you should try this formula against all known values to ensure there are no collisions.
If you don't care which value gets picked, change RANK() to ROW_NUMBER(). If you do care, do it anyway, but also add another term after the CASE statement in ORDER BY, separated by a comma, with the logic you want - for example if you want the first value alphabetically, do this:
...
ORDER BY CASE...END, "PQ"."Value")
...
I have around 40 queries which run on various tables and at the end of each query is a sub query for calculation of some metrics based on the values from above sub-queries. The final sub-query in each query looks like this -
SELECT dimension,
CASE when(impressions!=0) THEN clicks/impressions*100
ELSE NULL
END ctr,
CASE when(impressions!=0) THEN gross_rev/impressions*1000
ELSE NULL
END cpm,
CASE when(clicks!=0) THEN gross_rev/clicks
ELSE NULL
END cpc,
CASE when(actions!=0) THEN gross_rev/actions
ELSE NULL
END AS cpa,
CASE when(post_click!=0) THEN gross_rev/post_click
ELSE NULL
END AS pc_cpa
I want to reuse these formulas so wanted to write a function. But a vertica function does not return more than one columns.
Any idea on how to re-use these sub queries in a function or any other reusability technique for vertica.
You can do that by creating your own Vertica UDX , the UDX should be of TRANSFORM type , using this type of UDX the output schema is unrelated to the input schema , so basically you can provide list of columns as an input and the output can be different list with different size .
You can see some examples in GitHub https://github.com/vertica/Vertica-Extension-Packages
I hope you will find it useful
Thanks
This should be a simple one, but I have not found any solution:
The normal way is using an alias like this:
CASE WHEN ac_code='T' THEN 'time' ELSE 'purchase' END as alias
When using alias in conjunction with UNION ALL this causes problem because the alias is not treated the same way as the other columns.
Using an alias to assign the value is not working. It is still treated as alias, though it has the column name.
CASE WHEN ac_code='T' THEN 'time' ELSE 'purchase' END as ac_subject
I want to assign a value to a column based on a condition.
CASE WHEN ac_code='T' THEN ac_subject ='time' ELSE ac_subject='purchase' END
Now I get the error message
UNION types character varying and boolean cannot be matched
How can I assign a value to a column in a case statement without using an alias in the column (shared by other columns in UNION)?
Here is the whole (simplified) query:
SELECT hr_id,
CASE WHEN hr_subject='' THEN code_name ELSE hr_subject END
FROM hr
LEFT JOIN code ON code_id=hr_code
WHERE hr_job='123'
UNION ALL
SELECT po_id,
CASE WHEN po_subject='' THEN code_name ELSE po_subject END
FROM po
LEFT JOIN code ON code_id=po_code
WHERE po_job='123'
UNION ALL
SELECT ac_id,
CASE WHEN ac_code='T' THEN ac_subject='time' ELSE ac_subject='purchase' END
FROM ac
WHERE ac_job='123'
There is no alias in your presented query. You are confusing terms. This would be a column alias:
CASE WHEN hr_subject='' THEN code_name ELSE hr_subject END AS ac_subject
In a UNION query, the number of columns, column names and data types in the returned set are determined by the first row. All appended rows have to match the row type. Column names in appended rows (including aliases) are just noise and ignored. Maybe useful for documentation, nothing else.
The = operator does not assign anything in a SELECT query. It's the equality operator that returns a boolean value. TRUE if both operands are equal, etc. This returns a boolean value: ac_subject='time' Hence your error message:
UNION types character varying and boolean cannot be matched
The only way to "assign" a value to a particular output column in this query is to include it at the right position in the SELECT list.
The information in the question is incomplete, but I suspect you are also confusing the empty string ('') with the NULL value. A distinction that you need to understand before doing anything else with relational databases. Maybe start here. In this case you would rather use COALESCE to provide a default for NULL values:
SELECT hr_id, COALESCE(hr_subject, code_name) AS ac_subject
FROM hr
LEFT JOIN code ON code_id=hr_code
WHERE hr_job = '123'
UNION ALL
SELECT po_id, COALESCE(po_subject, code_name)
FROM po
LEFT JOIN code ON code_id=po_code
WHERE po_job = '123'
UNION ALL
SELECT ac_id, CASE WHEN ac_code = 'T' THEN 'time'::varchar ELSE 'purchase' END
FROM ac
WHERE ac_job = '123'
Just an educated guess, assuming type varchar. You should have added table qualification to column names to clarify their origin. Or table definitions to clarify everything.
The CASE expression is supposed to return a value, e.g. 'time'.
Your value is another expression subject ='time' which is a boolean (true or false).
Is this on purpose? Does the other query you glue with UNION have a boolean in that place, too? Probably not, and this is what the DBMS complains about.
I found the problem.
CASE WHEN hr_subject=’’ THEN code_name ELSE hr_subject END
The columns code_name and hr_subject was different length. This caused the unpredictable result. I think that aliases can work now.
Thank you for your support.
I want to update a column according to another column value.
for example, In Value column i have numbers between 0 to 1.
I want to check values and if:
Values < 0.45 set ValueStatus=Bad
Values >=0.45 and values<0.55 set ValueStatus =SoSo
Values >= 0.55 set ValueStatus=Good
I wrote the query like this:
update table
set ValueStatus=(case
when Values<'0.45' then 'Bad'
when (Values>='0.45' and Values<'0.55') then 'SoSo'
when Values>='0.55' then 'Good'
else Values
end)
But i get this error :
Error converting data type varchar to float.
Type of Values is Float and ValueStatus is Nvarchar(50)
Thanks
try this (you were adding ' to the numbers and SQL takes them as varchar) :
update table
set ValueStatus=(case when Values<0.45 then 'Bad'
when Values>=0.45 and Values<0.55 then 'SoSo' when Values>=0.55
then 'Good' else Values end )
I believe your problem is based on how the case statement determines the return type. You can read about it here and here.
The numeric types have a higher precedence than the string types. With the else values, you have four clauses in the `case. Three return strings; one returns a number. The number trumps the types so it tries to turn everything into a number.
You can mimic this problem with:
select (case when 1=1 then 'abc' else 12.3 end)
Happily, you can fix this by removing the else clause which is not needed in this case.