On 2017/06/21 21:37, Jeevan Ladhe wrote:
> Hi Amit,
>
> On Thu, Jun 15, 2017 at 12:31 PM, Amit Langote <
> Langote_Amit_f8@lab.ntt.co.jp> wrote:
>
>> Oops, I meant to send one more comment.
>>
>> On 2017/06/15 15:48, Amit Langote wrote:
>>> BTW, I noticed the following in 0002
>> + errmsg("there exists a default
>> partition for table \"%s\", cannot
>> add a new partition",
>>
>> This error message style seems novel to me. I'm not sure about the best
>> message text here, but maybe: "cannot add new partition to table \"%s\"
>> with default partition"
>>
>
> This sounds confusing to me, what about something like:
> "\"%s\" has a default partition, cannot add a new partition."
It's the comma inside the error message that suggests to me that it's a
style that I haven't seen elsewhere in the backend code. The primary
error message here is that the new partition cannot be created. "%s has
default partition" seems to me to belong in errdetail() (see "What Goes
Where" in [1].)
Or write the sentence such that the comma is not required. Anyway, we can
leave this for the committer to decide.
> Note that this comment belongs to patch 0002, and it will go away
> in case we are going to have extended functionality i.e. patch 0003,
> as in that patch we allow user to create a new partition even in the
> cases when there exists a default partition.
Oh, that'd be great. It's always better to get rid of the error
conditions that are hard to communicate to users. :) (Although, this
one's not that ambiguous.)
Thanks,
Amit
[1] https://www.postgresql.org/docs/devel/static/error-style-guide.html