-Re: Problem with rank() and dense_rank()
Lars Francke 2013-08-09, 15:44
Thank you very much guys for the response.
I saw your thread but I must have totally missed your resolution of
the issue. Sorry for the "repost".
I'd be happy to file an issue as well.
On Fri, Aug 9, 2013 at 4:26 PM, j.barrett Strausser
<[EMAIL PROTECTED]> wrote:
> rank is case-sensitive. Use the lower-case invocation.
> It is a rather odd error for the underlying cause and one I would have never
> thought to check.
> I ran across the same issue you did. It was discussed in the email thread I
> started a few weeks ago titled "Semantics of Rank".
> I'd advise you to go look in the archives for that thread.
> On Fri, Aug 9, 2013 at 9:39 AM, Lars Francke <[EMAIL PROTECTED]> wrote:
>> Hi again,
>> CREATE TABLE test (a INT);
>> DENSE_RANK() OVER (PARTITION BY a),
>> FROM test;
>> this minimal example fails with:
>> FAILED: SemanticException Failed to breakup Windowing invocations into
>> Groups. At least 1 group must only depend on input columns. Also check
>> for circular dependencies.
>> Underlying error:
>> org.apache.hadoop.hive.ql.exec.UDFArgumentTypeException: One or more
>> arguments are expected.
>> Does anyone have an idea if we're using this wrong? It is correct that
>> RANK() and DENSE_RANK() take a parameter (by looking at the code). I'm
>> not entirely sure what it does but in any case it's undocumented and
>> the tests seem to work without one too.
>> Thank you for your help.