On Tue, Nov 24, 2015 at 9:31 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Tue, Nov 24, 2015 at 1:06 AM, Amit Langote
> <Langote_Amit_f8@lab.ntt.co.jp> wrote:
>> While going through nodeGather.c, I noticed portions of the file header
>> comment that may have been obsoleted by recent revisions of the relevant
>> parellelism code. For example, there is a reference to PartialSeqScan node
>> which did not make it into the tree. Attached fixes it. Also, wondering if
>> the semantics of Gather node is that of Scan or more generic Plan? That is
>> to ask whether the following edit makes sense:
>>
>> * nodeGather.c
>> - * Support routines for scanning a plan via multiple workers.
>> + * Support routines for getting the result from a plan via multiple
>> + * workers.
>> *
>
> Well I think "scanning a plan" is clear enough even if it's
> technically a Scan. But I agree the second change is needed.
Err, even if it's NOT technically a scan.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
