On Sun, Sep 18, 2022 at 5:08 PM Peter Geoghegan <pg@bowt.ie> wrote:
> That was left behind following a recent rebase. Will fix.
Attached revision fixes this issue, plus the
ExecIncrementalSortInitializeWorker() issue.
It also adds a lot more fixes which were missed earlier because I
didn't use "strict" when running clang-tidy from the command line
(just in my editor). This includes a fix for the mdexists() issue that
you highlighted.
I expect that this patch series will bitrot frequently, so there is no
good reason to not post revisions frequently.
The general structure of the patchset is now a little more worked out.
Although it's still not close to being commitable, it should give you
a better idea of the kind of structure that I'm aiming for. I think
that this should be broken into a few different parts based on the
area of the codebase affected (not the type of check used). Even that
aspect needs more work, because there is still one massive patch --
this is now the sixth and final patch.
It seems like a good idea to at least have separate commits for both
the regex code and the timezone code, since these are "quasi-vendored"
areas of the code base.
--
Peter Geoghegan