Closed gregsdennis closed 1 year ago
The PR needs to generalise the rules which apply to existence tests and comparisons.
I don't follow. How does the current text not do this?
This will necessarily reintroduce certain conversions...
I'm not sure it does. We've been careful to explicitly declare the behavior of everything where it's valid/well-formed. (Personally I think the conversions are clearer, but that's not what this PR is about.)
Argh! I dismissed my review but this ended up approving the PR rather than changing it to a comment. (The github docs claim is would change to a comment.) @cabo & @goessner: please note.
I'd also like to revisit the text quoted in https://github.com/ietf-wg-jsonpath/draft-ietf-jsonpath-base/pull/418#discussion_r1127034954 to add the empty nodelist -> Nothing
case that is notably missing. A separate PR, though, and maybe only if this goes through.
Builds on #412. Attempts to address this comment from @cabo:
Only valid/well-formed scenarios are explained; others are considered invalid/poorly-formed.