Question

I need to build a Twitter module entirely on the client side.

I intended to use the Twitter API for the data and YQL to get across Same Origin Policy, and also for its higher rate limiting. I have used this combination before with success.

One thing I need to do is to get a list of tweets with a hashtag and from a finite set of users.

This gets tweets with the #stackoverflow hash tag by users alexdickson and lizardbill.

I also need to get a different set of tweets.

This gets all tweets with the #stackoverflow hash tag.

Now, I need to display the second collection but excluding the results from the first set. This would be a diff of the results, like PHP’s array_diff().

For completeness, I may also need to achieve something by getting an intersection of two results, that is, a list of tweets which appear in both result sets. This is similar to PHP’s array_intersect().

It has been a while since I’ve used SQL and can’t think of how to achieve this in YQL.

Answer 1

YQL supports sub-selects (subqueries). See Joining Tables with Sub-selects. Sub-selects are used with the IN operator, to filter the results of the outer select.

For your particular query, there is also a twitter.search Community Open Data Table available to save you from building the Twitter URLs and using the json table.

select *
from twitter.search
where q="#stackoverflow" 
and id not in (
    select id 
    from twitter.search 
    where q="#stackoverflow from:alexdickson, OR from:lizardbill"
)

^{(Try this query in the YQL console.)}