lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Miller <markrmil...@gmail.com>
Subject Re: TopDocCollector limits
Date Wed, 30 Sep 2009 17:47:45 GMT
Way the heck better - Hits is horrible for that. It caches like 100 hits
and then keeps searching when you exhaust the cache (been I while since
I've looked at the exact numbers). Its horribly inefficient for checking
every hit.

Hits will end up using a Collector anyway - and then throw a speed trap
for what you want on top of it. Part of why we deprecated it. To much
misuse.

Max Lynch wrote:
> Thanks Mark that's exactly what I need.  How does the performance of
> processing each document in the collect method of HitCollector compare to
> looping through the Hits in the deprecated Hits class?
>
> On Tue, Sep 29, 2009 at 7:40 PM, Mark Miller <markrmiller@gmail.com> wrote:
>
>   
>> Max Lynch wrote:
>>     
>>> Hi,
>>> I am developing a search system that doesn't do pagination (searches are
>>>       
>> run
>>     
>>> in the background and machine analyzed).  However, TopDocCollector makes
>>>       
>> me
>>     
>>> put a limit on how many results I want back.  For my system, each result
>>> found is important.  How can I make it collect every result found?
>>>
>>> Thanks.
>>>
>>>
>>>       
>> You would want to use a Collector in 2.9 or HitCollector < 2.9 most likely.
>>
>> --
>> - Mark
>>
>> http://www.lucidimagination.com
>>
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: java-user-help@lucene.apache.org
>>
>>
>>     
>
>   


-- 
- Mark

http://www.lucidimagination.com




---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org


Mime
View raw message