Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hello,
I have set analysis. I want to choose all data exept some data.I mean something like "ID < 7.77e+014 and ID > 7.99e+014".
I tryed many differnt ways
Count({$<ID={"=ID<=7.7700001678389e+014" },ID={"ID>=7.7700002244059e+014">}surveyID)-------------Nothing
Count({$<ID={"7.77*"}>}surveyID)-----------------------select exactly that I want not includ.
Count({<ID={"=ID<=7.77e+014"}>}surveyID)--------------it works but without second part of condition.
I don't know why it loads my ID in this format 7.7700002244059e+014. It should be 777000022440585.
Thanks for your help,
Vladimir
7.7799999999999e+014 is way larger than 8,9,10,11 (it is in scientific notation 7.7799 * 10^14), so I think the scientific notation don't get correctly parsed in the set element
(edit: I just tried, and it works at my place, you I don't know why you get "0,1,2,3,4 and 8,9,10,11, ..." returned. This doesn't make sense to me)
Anyway, have you had a chance to look into my above sample and have you tried using the evaluate function in your load?
I think your expression could then look similar to
count({$<ID={"=ID>777000000000000 and ID<799999999999999"} >} Value)
or
count({$<ID={"=ID>7.77e14 and ID<7.99999999e+014"} >} Value)
Regards,
Stefan
Yeah, that's an interesting question.
There are some, but not too many posts here in the forum which talk about problems reading in large (integer) values with more than some 14 digits.
One posted solution is to use evaluate() on the field with the large integer number. Indeed, this did the job for me with QV10 SR3, like
LOAD *, evaluate(ID) as ID2 INLINE [
ID, Value
777000022440585, 1
777000022440586, 2
12321321, 3
877000022440585, 4
7.7700002244059e+014, 5
6.7700002244059e+014, 6
8.7700002244059e+014,7
1234567890123456, 8
];
Then you could use something like
=sum({$<ID2={"=ID2<>777000022440585"} >} Value)
to filter out the value you want to exclude.
See also attached,
Stefan
for this case: when Id <5 and Id > 7. I found out
Count({$<ID={"=ID<7.77e+014","=ID >7.7799999999999e+014"}>}ID)-----------------------works perfect.
BUT
When ID>5 and ID<7 I can't find right solution.
Maybe someone knows how to exclud 5,6,7?
Thanks,
To exclude values, try
Count({$<ID -= {5,6,7}>} ID)
Please note the minus-equal sign.
Something like this should combine it with your previous expression:
Count({$<ID={"=ID<7.77e+014","=ID >7.7799999999999e+014"} - {5,6,7} >} ID)
Hi swuehl,
Count({$<ID={"=ID<7.77e+014","=ID >7.7799999999999e+014"}>}ID)------
This expressions selects like example 0,1,2,3,4 and 8,9,10,11, ...
I need expressions which select 5,6,7 only. In my case it's thousands. I can't select them one by one .
I don't understandt how to group them like from 5 to 7.
Maybe it's use to use condition IF
This my expressions
Count({$<SOURCE={'2'},TASK_COMPLETION={Yes}>} TASK_COMPLETION)/
Count({$<SOURCE={'2'}>}TOTAL <Period> TASK_COMPLETION)
Instead of SOURCE=2 should be ID from 7.77e+014 to 7.99999999e+014.
Who knows how to write expression?
Thanks,
7.7799999999999e+014 is way larger than 8,9,10,11 (it is in scientific notation 7.7799 * 10^14), so I think the scientific notation don't get correctly parsed in the set element
(edit: I just tried, and it works at my place, you I don't know why you get "0,1,2,3,4 and 8,9,10,11, ..." returned. This doesn't make sense to me)
Anyway, have you had a chance to look into my above sample and have you tried using the evaluate function in your load?
I think your expression could then look similar to
count({$<ID={"=ID>777000000000000 and ID<799999999999999"} >} Value)
or
count({$<ID={"=ID>7.77e14 and ID<7.99999999e+014"} >} Value)
Regards,
Stefan
Thanks Stefan,
You are right.
Count({$<ID={"=ID>=7.77e+014 AND ID <=7.7799999999999e+014"}>}ID)
It works !!!!!!!!!