Robert,
On Tuesday, June 17, 2014, Robert Haas <
robertmhaas@gmail.com> wrote:
After sending that one (1) email, I was promptly told that "I'm very
disappointed to hear that the mechanical pieces around making RLS easy
for users to use ... is receiving such push-back." The push-back, at
that point in time, consisted of one (1) email. Several more emails
have been sent that time, including the above-quoted text, seeming to
me to imply that the people who are concerned about this feature are
being unreasonable. I don't believe I am the only such person,
although I may be the main one right at the moment, and you may not be
entirely surprised to hear that I don't think I'm being unreasonable.
I'm on my phone at the moment but that looks like a quote from me. My email and concern there was regarding the specific suggestion that we could check off the "RLS" capability which users have been asking us to provide nearly since I started with PG by saying that they could use Updatable SB views. I did not intend it as a comment regarding the specific technical concerns raised and have been responding to and trying to address those independently and openly.
I've expressed elsewhere on this thread my gratitude that the technical concerns are being brought up now, near the beginning of the cycle, so we can address them. I've been working with others who are interested in RLS on a wiki page to outline and understand the options and identify dependencies and priorities. Hopefully the link will be posted shortly (again, not at a computer right now) and we can get comments back. There are some very specific questions which really need to be addressed and which I've mentioned before (in particular the question of what user the functions in a view definition should run as, both for "normal" views, for SB views, and for when an RLS qual is included and run through that framework, and if doing so would address some of the concerns which have been raised regarding selects running code).
Thanks,
Stephen
