On Thu, Mar 19, 2020 at 3:55 AM Chris Bandy <bandy.chris@gmail.com> wrote:
>
>
> Sorry for these troubles. Attached are patches created using `git
> format-patch -n -v6` on master at 487e9861d0.
>
No problem. I have extracted your code changes as a separate patch
(see attached) as I am not sure we want to add tests for these cases.
This doesn't apply in back-branches, but I think that is small work
and we can do that if required. The real question is do we want to
back-patch this? Basically, this improves the errors in certain cases
by providing additional information that otherwise the user might need
to extract from error messages. So, there doesn't seem to be pressing
need to back-patch this but OTOH, we have mentioned in docs that we
support to display this information for all SQLSTATE class 23
(integrity constraint violation) errors which is not true as we forgot
to adhere to that in some parts of code.
What do you think? Anybody else has an opinion on whether to
back-patch this or not?
[1] - https://www.postgresql.org/docs/devel/errcodes-appendix.html
--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com