Robert Haas <robertmhaas@gmail.com> writes:
> On Thu, Mar 9, 2017 at 6:45 PM, Neha Khatri <nehakhatri5@gmail.com> wrote:
>> With this, if an installcheck is done, that might also have been done with
>> the expectation that the output will be in 'escape' format. In that case,
>> how much is it justified to hard code the format for regression database?
>> However, I agree that there are not many bytea outputs in the current
>> regression suite
> I don't understand this. People don't run the regression tests to get
> the output. They run the regression tests to see whether they pass.
> While it may not be possible to make them pass with arbitrarily-crazy
> settings, that's not a reason not to patch up the cases we can handle
> sanely.
I think the question that has to be settled to move this forward is
whether we're content with hard-wiring something for bytea_output
(as in Jeff's installcheck_bytea_fix_2.patch, which I concur with
Peter is superior to installcheck_bytea_fix_1.patch), or whether
we want to hold out for a more flexible solution, probably about like
what I sketched in
https://www.postgresql.org/message-id/30246.1487202663%40sss.pgh.pa.us
I think the more flexible solution is definitely a desirable place to
get to, but somehow I doubt it's going to happen for v10. Meanwhile
the question is whether adding more hard-wired behavior in pg_regress
is desirable or not.
I tend to vote with Andres that it's not worth the trouble, but
considering that it's only a 2-line change, I won't stand in the
way if some other committer is convinced this is an improvement.
regards, tom lane
