Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Pig, mail # user - COALESCE UDF?


Copy link to this message
-
Re: COALESCE UDF?
Russell Jurney 2013-09-04, 16:32
On the other hand, if you use a java UDF, consider contributing it.

Sent from my iPhone

On Sep 4, 2013, at 9:10 AM, Serega Sheypak <[EMAIL PROTECTED]> wrote:

> Use simple jython UDF. Its 3 lines of code
> 04.09.2013 18:04 пользователь "Sajid Raza" <[EMAIL PROTECTED]> написал:
>
>> For my two cents' worth I agree, it's nicer to have coalesce than the
>> conditional operator.
>>
>> On Sep 4, 2013, at 8:50 AM, Something Something <[EMAIL PROTECTED]>
>> wrote:
>>
>>> What if you've 10 fields?
>>>
>>>
>>> On Wed, Sep 4, 2013 at 4:30 AM, Ruslan Al-Fakikh <[EMAIL PROTECTED]
>>> wrote:
>>>
>>>> Hi,
>>>>
>>>> I think you could mimic it with an expression like this:
>>>> b = foreach a generate ((field1 is null) ? ((field2 is null) ? null :
>>>> field2) : field1);
>>>>
>>>> Hope that helps,
>>>> Ruslan
>>>>
>>>>
>>>> On Wed, Sep 4, 2013 at 9:50 AM, Something Something <
>>>> [EMAIL PROTECTED]> wrote:
>>>>
>>>>> Is there a UDF in Piggybank (or somewhere) that will mimic
>> functionality
>>>> of
>>>>> the COALESCE function in MySQL:
>> http://www.w3resource.com/mysql/comparision-functions-and-operators/coalesce-function.php
>>>>>
>>>>> I know it will be very (very) easy to write this, but just don't want
>> to
>>>>> create one if one already exists.  I looked but couldn't find one.
>>>>>
>>>>> Please let me know.  Thanks.
>>