# Active record object for reuse

**URL:** <https://www.ruby-forum.com/t/active-record-object-for-reuse/137544>\
**Category:** Rails\
**Created:** [May 14, 2008, 3:25am UTC](https://www.ruby-forum.com/t/active-record-object-for-reuse/137544 "2008-05-14T03:25:33Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![BENI](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/b/bc8723/32.png) [@BENI](https://www.ruby-forum.com/u/BENI)\
**Post date:** [May 14, 2008, 3:25am UTC](https://www.ruby-forum.com/t/active-record-object-for-reuse/137544/1 "2008-05-14T03:25:33Z")

</div>

hI,

I have a fairly complex mysql query which span through ~500000 records  
and multiple tables …which take almost 30 secs to run and generates  
~5000 rows .

The query has been optimised.

Now … i want to do the foll with this

1\> Paginate …I do not want to rerun the query with LIMIT/OFFET  
clause etc. The object obtained in the first place.Can it be some how  
stored and reused?

2\>Sort on different combination of columns …here also I want to use  
ruby and do all the functionality.

But how to save the active record object and reuse it .

Basically I want to avoid the 30 second delay for each query.

Cheers

Rajib

---

<div class="post-metadata">

**Author:** ![BENI](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/b/bc8723/32.png) [@BENI](https://www.ruby-forum.com/u/BENI)\
**Post date:** [May 14, 2008, 10:22am UTC](https://www.ruby-forum.com/t/active-record-object-for-reuse/137544/2 "2008-05-14T10:22:15Z")

</div>

On May 14, 2:25 am, BENI [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> 1\> Paginate …I do not want to rerun the query with LIMIT/OFFET  
> clause etc. The object obtained in the first place.Can it be some how  
> stored and reused?
> 
> 2\>Sort on different combination of columns …here also I want to use  
> ruby and do all the functionality.

1. optimise it some more, use explain select …, make sure you  
understand what is going on. Maybe you can structure you data so that  
the query isn’t so complicated
2. cache it (memcached etc…)

Fred

---

<div class="post-metadata">

**Author:** ![BENI](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/b/bc8723/32.png) [@BENI](https://www.ruby-forum.com/u/BENI)\
**Post date:** [May 14, 2008, 3:03pm UTC](https://www.ruby-forum.com/t/active-record-object-for-reuse/137544/3 "2008-05-14T15:03:02Z")

</div>

Also, that sounds like a very expensive query to leave in the hands of  
indiscriminate users. It’s such a costly query it sounds like there  
should be another way to do it. Any chance it’s a once-per-day type  
job that you could off load to a background process or cron job?  
Another possibility would be warehousing the data (e.g.,  
ActiveWarehouse, if the project is still active) and supporting  
queries only on data that has been warehoused.

On May 14, 4:21 am, Frederick C. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)