Horton

Re: remaining sql/json patches

amit <amitlangote09@gmail.com>

View thread
From: Amit Langote <amitlangote09@gmail.com>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
Cc: Erik Rijkers <er@xs4all.nl>, jian he <jian.universality@gmail.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Andrew Dunstan <andrew@dunslane.net>
Date: 2023-07-20T08:19:02Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. SQL/JSON: Various improvements to SQL/JSON query function docs

  2. SQL/JSON: Fix some obsolete comments.

  3. SQL/JSON: Fix issues with DEFAULT .. ON ERROR / EMPTY

  4. JSON_TABLE: Add support for NESTED paths and columns

  5. Fix JsonExpr deparsing to emit QUOTES and WRAPPER correctly

  6. Fix typo introduced in 6185c9737

  7. Add basic JSON_TABLE() functionality

  8. Avoid splitting errmsg string to span multiple lines

  9. Add SQL/JSON query functions

  10. Implement various jsonpath methods

  11. Add soft error handling to some expression nodes

  12. Adjust populate_record_field() to handle errors softly

  13. Refactor code used by jsonpath executor to fetch variables

  14. Test EXPLAIN (FORMAT JSON) ... XMLTABLE

  15. Simplify productions for FORMAT JSON [ ENCODING name ]

  16. Add trailing commas to enum definitions

  17. doc: add missing <returnvalue> and whitespace

  18. Add more SQL/JSON constructor functions

  19. Rename a nonterminal used in SQL/JSON grammar

  20. Some refactoring to export json(b) conversion functions

  21. Don't include CaseTestExpr in JsonValueExpr.formatted_expr

  22. Code review for commit b6e1157e7d

  23. Pass constructName to transformJsonValueExpr()

  24. Unify JSON categorize type API and export for external use

  25. Make some indentation in gram.y consistent

  26. Allow most keywords to be used as column labels without requiring AS.

  27. Reduce size of backend scanner's tables.

  28. Use perfect hashing, instead of binary search, for keyword lookup.

Attachments

On Wed, Jul 19, 2023 at 5:17 PM Amit Langote <amitlangote09@gmail.com> wrote:
> On Wed, Jul 19, 2023 at 12:53 AM Alvaro Herrera <alvherre@alvh.no-ip.org> wrote:
> > On 2023-Jul-18, Amit Langote wrote:
> >
> > > Attached updated patches.  In 0002, I removed the mention of the
> > > RETURNING clause in the JSON(), JSON_SCALAR() documentation, which I
> > > had forgotten to do in the last version which removed its support in
> > > code.
> >
> > > I think 0001 looks ready to go.  Alvaro?
> >
> > It looks reasonable to me.
>
> Thanks for taking another look.
>
> I will push this tomorrow.

Pushed.

> > > Also, I've been wondering if it isn't too late to apply the following
> > > to v16 too, so as to make the code look similar in both branches:
> >
> > Hmm.
> >
> > > 785480c953 Pass constructName to transformJsonValueExpr()
> >
> > I think 785480c953 can easily be considered a bugfix on 7081ac46ace8, so
> > I agree it's better to apply it to 16.
>
> OK.

Pushed to 16.

> > > b6e1157e7d Don't include CaseTestExpr in JsonValueExpr.formatted_expr
> >
> > I feel a bit uneasy about this one.  It seems to assume that
> > formatted_expr is always set, but at the same time it's not obvious that
> > it is.  (Maybe this aspect just needs some more commentary).
>
> Hmm, I agree that the comments about formatted_expr could be improved
> further, for which I propose the attached.  Actually, staring some
> more at this, I'm inclined to change makeJsonValueExpr() to allow
> callers to pass it the finished 'formatted_expr' rather than set it by
> themselves.
>
> >  I agree
> > that it would be better to make both branches identical, because if
> > there's a problem, we are better equipped to get a fix done to both.
> >
> > As for the removal of makeCaseTestExpr(), I agree -- of the six callers
> > of makeNode(CastTestExpr), only two of them would be able to use the new
> > function, so it doesn't look of general enough usefulness.
>
> OK, so you agree with back-patching this one too, though perhaps only
> after applying something like the aforementioned patch.

I looked at this some more and concluded that it's fine to think that
all JsonValueExpr nodes leaving the parser have their formatted_expr
set.  I've updated the commentary some more in the patch attached as
0001.

Rebased SQL/JSON patches also attached.  I've fixed the JSON_TABLE
syntax synopsis in the documentation as mentioned in my other email.

--
Thanks, Amit Langote
EDB: http://www.enterprisedb.com