Hi, Sorry, I have missed this in the original patch. I am surprised how that happened. But thanks for catching the same and fixing it.
The changes are straightforward and look good to me. However, I have kept the existing test of an empty array as is, assuming that it is one of the valid tests. It now returns zero rows instead of an error. Your added test case covers this issue. Thanks On Fri, Jun 14, 2024 at 9:34 PM David E. Wheeler <da...@justatheory.com> wrote: > > > > On Jun 14, 2024, at 11:25, Chapman Flack <jcfl...@acm.org> wrote: > > > > I would s/extepsions/exceptions/ in the added documentation. :) > > Bah, fixed and attached, thanks. > > > Offhand (as GitHub PRs aren't really The PG Way), if they were The Way, > > I would find this one a little hard to follow, being based at a point > > 28 unrelated commits ahead of the ref it's opened against. I suspect > > 'master' on theory/postgres could be fast-forwarded to f1affb6 and then > > the PR would look much more approachable. > > Yeah, I pushed the PR and branch before I synced master, and GitHub was > taking a while to notice and update the PR. I fixed it with `git commit > --all --amend --date now --reedit-message HEAD` and force-pushed (then > fixed the typo and fixed again). > > D > > > > -- *Jeevan Chalke* *Principal, ManagerProduct Development* enterprisedb.com <https://www.enterprisedb.com>
v3-0001-Teach-jsonpath-string-to-unwrap-in-lax-mode.patch
Description: Binary data
