Tom,
* Tom Lane (tgl@sss.pgh.pa.us) wrote:
> broken that way as rowsecurity.sql, which is (still) creating roles
> named "alice" and "bob", but it's not acceptable like this.
Attached is a patch to fix all of the role usage in rowsecurity.sql
(I believe, feel free to let me know if there's anything else). In
particular, all of the roles are changed to begin with 'regress_', and
they are all now created with NOLOGIN.
I'll plan to push this in the next day or so.
I'll also work up a patch for the rest of the CREATE USER/ROLE usage in
the core regression tests.
This will result in a bit of not-strictly-necessary code churn, but
having consistency in the code base really is valuable where we have an
agreed upon policy as to what all the tests should be doing, for new
(and even old) developers to follow.
Comments and concerns welcome, of course.
Thanks!
Stephen
