On 5/8/18 8:11 AM, Stephen Frost wrote:
> Greetings,
>
> * Michael Paquier (michael@paquier.xyz) wrote:
>> On Sun, May 06, 2018 at 09:14:06PM -0400, Stephen Frost wrote:
>>> While I appreciate the support, I'm not sure that you're actually
>>> agreeing with me.. I was arguing that braces should be on their own
>>> line and therefore there would be a new line for the brace.
>>> Specifically, when moving lines between hashes, it's annoying to have to
>>> also worry about if the line being copied/moved has braces at the end or
>>> not- much easier if they don't and the braces are on their own line.
>>
>> I should have read that twice. Yes we are not on the same line. Even
>> if a brace is on a different line, per your argument it would still be
>> nicer to add a comma at the end of each last element of a hash or an
>> array, which is what you have done in the tests of pg_dump, but not
>> something that the proposed patch does consistently. If the formatting
>> is automated, the way chosen does not matter much, but the extra last
>> comma should be consistently present as well?
>
> Yes, that would be nice as well, as you'd be able to move entries around
> more easily that way.
I'm a fan of the final comma as it makes diffs less noisy.
Regards,
--
-David
david@pgmasters.net