I'm trying to set up a document partitioned index that can handle a ranges of terms or wildcards for queries.

So, if instead of querying "the" AND "green" AND "goblin", it could handle "the" AND "green" AND "go*" (which would also return "goddess", for instance). Or a search that used "the" AND "d"-"f" AND "goblin", handling all values between "d" and "f".

Using a typical document-partitioned index, I'm guessing that you might first resolve the wildcard into a list of terms, and then do a query in the normal fashion. However, this seems rather inefficient. Is there a separate data structure that would be recommended to handle this sort of additional functionality?

Thanks,
David

You could store n-grams of terms, to support some limited wildcard
searching.

--
Christopher L Tubbs II
http://gravatar.com/ctubbsii
On Tue, Jan 22, 2013 at 12:13 PM, Slater, David M.
<[EMAIL PROTECTED]>wrote:

> I’m trying to set up a document partitioned index that can handle a ranges
> of terms or wildcards for queries.****
>
> ** **
>
> So, if instead of querying “the” AND “green” AND “goblin”, it could handle
> “the” AND “green” AND “go*” (which would also return “goddess”, for
> instance). Or a search that used “the” AND “d”-“f” AND “goblin”, handling
> all values between “d” and “f”.****
>
> ** **
>
> Using a typical document-partitioned index, I’m guessing that you might
> first resolve the wildcard into a list of terms, and then do a query in the
> normal fashion. However, this seems rather inefficient. Is there a separate
> data structure that would be recommended to handle this sort of additional
> functionality?****
>
> ** **
>
> Thanks,
> David****
>

You can use a modified intersecting iterator which matches on the prefix go
instead of doing a full match.
On Tue, Jan 22, 2013 at 3:40 PM, Christopher <[EMAIL PROTECTED]> wrote:

> You could store n-grams of terms, to support some limited wildcard
> searching.
>
>
>
> --
> Christopher L Tubbs II
> http://gravatar.com/ctubbsii
>
>
> On Tue, Jan 22, 2013 at 12:13 PM, Slater, David M. <
> [EMAIL PROTECTED]> wrote:
>
>> I’m trying to set up a document partitioned index that can handle a
>> ranges of terms or wildcards for queries.****
>>
>> ** **
>>
>> So, if instead of querying “the” AND “green” AND “goblin”, it could
>> handle “the” AND “green” AND “go*” (which would also return “goddess”, for
>> instance). Or a search that used “the” AND “d”-“f” AND “goblin”, handling
>> all values between “d” and “f”.****
>>
>> ** **
>>
>> Using a typical document-partitioned index, I’m guessing that you might
>> first resolve the wildcard into a list of terms, and then do a query in the
>> normal fashion. However, this seems rather inefficient. Is there a separate
>> data structure that would be recommended to handle this sort of additional
>> functionality?****
>>
>> ** **
>>
>> Thanks,
>> David****
>>
>
>