Re: Bypassing NULL elements in row_to_json function

On 04/08/2016 08:31 AM, Michael Nolan wrote:
> I'm looking at the possibility of using JSON as a data exchange format
> with some apps running on both PCs and Macs.   .
>
> The table I would be exporting has a lot of NULL values in it.  Is
> there any way to skip the NULL values in the row_to_json function and
> include only the fields that are non-null?

I guess it depends on your data.

Are the NULLs all in one field or scattered across fields?

Imagining this scenario:

fld_1     fld_2    fld_3
'val1_1'  NULL     'val1_3
NULL      'val2_2' 'val2_3'
'val3_3'  'val3_2'  NULL

How do you deal with the holes(NULL) on the receiving end?


> --
> Mike Nolan
> nolan@tssi.com
>
>


--
Adrian Klaver
adrian.klaver@aklaver.com