> On 14 Jun 2023, at 13:02, Andrew Dunstan <andrew@dunslane.net> wrote:
> On 2023-06-12 Mo 19:43, Nathan Bossart wrote:
>> On Tue, Jun 13, 2023 at 08:29:19AM +0900, Michael Paquier wrote:
>>
>>> I am actually a bit confused with the return value of
>>> CreateRestrictedProcess() on failures in restricted_token.c. Wouldn't
>>> it be cleaner to return INVALID_HANDLE_VALUE rather than 0 in these
>>> cases?
>>>
>> My suspicion is that this was chosen to align with CreateProcess and to
>> allow things like
>>
>> if (!CreateRestrictedProcess(...))
>
> Probably, it's been a while. I doubt it's worth changing at this point, and we could just change pg_regress.c to use
aboolean test like the above.
Done that way and pushed, thanks!
--
Daniel Gustafsson