"Jonah H. Harris" <jonah.harris@gmail.com> writes:
> While working on the join elimination patch, I was going through the
> trigger code and found quite a bit of nastiness in regard to naming
> and variable repurposing related to the addition of replication roles
> in 8.3. The most obvious issue is that tgenabled was switched from a
> bool to char to support replication roles. From a naming standpoint,
> the term "enabled" generally implies boolean and is fairly
> consistently used as such in other functions within the core. My
> initial preference would be to return tgenabled to its original
> boolean for use only in enabling/disabling triggers.
It would have been useful to make this criticism before 8.3 was
released. I don't think it's reasonable to change it now.
regards, tom lane
