On 2019-Jul-24, Ian Barwick wrote:
> It'd be better if such a hypothetical option validated the provided
> slot name anwyay, to prevent later surprises.
Hmm, but what would we do if the validation failed?
> Revised patch attached, which as Alvaro suggests removes the escaping
> and adds a comment explaining why the raw value can be passed as-is.
Heh, yesterday I revised the original patch as attached and was about to
push when the bell rang. I like this one because it keeps the comment
to one line and it mentions the function name in charge of the
validation (useful for grepping later on). It's a bit laconic because
of the long function name and the desire to keep it to one line, but it
seems sufficient to me.
BTW upper case letters are not allowed :-)
--
Álvaro Herrera https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
