Andrew Dunstan <andrew@dunslane.net> writes:
> Roger Leigh wrote:
>> Here we just force the locale to C. This does have the disadvantage
>> that --no-locale is made redundant, and any tests which are dependent
>> upon locale (if any?) will be run in the C locale.
> That is not a solution.
Right. I think you may have missed the point of what Peter was saying:
it's okay to force the locale to C on the *client* side of the tests.
The trick is to not change the environment that the temp server is
started with. This will take some redesign inside pg_regress, I think.
Probably initialize_environment needs to be split into two functions,
one that's run before firing off the temp server and one that runs
after.
regards, tom lane