I'm ok with the count returned being some estimate.  Say in this simple 
example if it returned 1 for just Joe, or 3 for John, Joe, and Jack that 
would be ok too.  I am also ok with restructuring my data in any way to 
more efficiently get this number.  

You mentioned creating a reference count document.  How would that look?  1 
doc per unique author, with a count of the total number of books he wrote 
so then I can do a range aggregation on that number?  What if I wanted to 
find "the number of authors who have written between 2-3 books that have a 
title containing E, F, H, or I" (still 2 in this case, John and Joe) ?  



On Thursday, June 19, 2014 6:43:41 PM UTC-4, Itamar Syn-Hershko wrote:
>
> This is a Map/Reduce operation, you'll be better off maintaining a 
> ref-count document IMO then trying to hack the aggregations framework to 
> support this
>
> Another reason for doing it that way is in a distributed environment some 
> aggregations can't be computed to an exact value - the Terms bucketing is 
> one example. So if you need exact values, I'd go for a model that does it.
>
> --
>
> Itamar Syn-Hershko
> http://code972.com | @synhershko <https://twitter.com/synhershko>
> Freelance Developer & Consultant
> Author of RavenDB in Action <http://manning.com/synhershko/>
>
>
> On Fri, Jun 20, 2014 at 1:34 AM, Mike <[email protected] <javascript:>> 
> wrote:
>
>> Assume each document is a book:  
>> { title: "A", author: "Mike" }
>> { title: "B", author: "Mike" }
>> { title: "C", author: "Mike" }
>> { title: "D", author: "Mike" }
>>
>> { title: "E", author: "John" }
>> { title: "F", author: "John" }
>> { title: "G", author: "John" }
>>
>> { title: "H", author: "Joe" }
>> { title: "I", author: "Joe" }
>>
>> { title: "J", author: "Jack" }
>>
>>
>> What is the best way to fin the number of authors who have written 
>> between 2-3 books?  In this case it would be 2, John and Joe.
>>
>> I know I can do a terms aggregation on author, set size to be very very 
>> large, and then on the client side traverse through the thousands of 
>> authors and count how many had between 2-3.  Is there a more efficient way 
>> to do this?  The cardinality aggregation is almost what I want, if only I 
>> could specify a min and max term count. 
>>
>>
>>  -- 
>> You received this message because you are subscribed to the Google Groups 
>> "elasticsearch" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/elasticsearch/22fc4e6d-bcac-426c-a343-ff1d36fc25de%40googlegroups.com
>>  
>> <https://groups.google.com/d/msgid/elasticsearch/22fc4e6d-bcac-426c-a343-ff1d36fc25de%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/2cab8d84-7c65-4f6e-ab39-3e2a0e859a87%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to