Re: remaining sql/json patches

jian he <jian.universality@gmail.com>

From: jian he <jian.universality@gmail.com>
To: Amit Langote <amitlangote09@gmail.com>
Cc: Tomas Vondra <tomas.vondra@enterprisedb.com>, Himanshu Upadhyaya <upadhyaya.himanshu@gmail.com>, Alvaro Herrera <alvherre@alvh.no-ip.org>, Erik Rijkers <er@xs4all.nl>, Andres Freund <andres@anarazel.de>, Andrew Dunstan <andrew@dunslane.net>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2024-03-27T03:41:50Z
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.

On Tue, Mar 26, 2024 at 6:16 PM jian he <jian.universality@gmail.com> wrote:
>
> On Fri, Mar 22, 2024 at 12:08 AM Amit Langote <amitlangote09@gmail.com> wrote:
> >
> > On Wed, Mar 20, 2024 at 9:53 PM Amit Langote <amitlangote09@gmail.com> wrote:
> > > I'll push 0001 tomorrow.
> >
> > Pushed that one.  Here's the remaining JSON_TABLE() patch.
> >

hi.
I don't fully understand all the code in json_table patch.
maybe we can split it into several patches, like:
* no nested json_table_column.
* nested json_table_column, with PLAN DEFAULT
* nested json_table_column, with PLAN ( json_table_plan )

i can understand the "no nested json_table_column" part,
which seems to be how oracle[1] implemented it.
I think we can make the "no nested json_table_column" part into  v17.
i am not sure about other complex parts.
lack of comment, makes it kind of hard to fully understand.

[1] https://docs.oracle.com/en/database/oracle/oracle-database/19/sqlrf/img_text/json_table.html



+/* Reset context item of a scan, execute JSON path and reset a scan */
+static void
+JsonTableResetContextItem(JsonTableScanState *scan, Datum item)
+{
+ MemoryContext oldcxt;
+ JsonPathExecResult res;
+ Jsonb   *js = (Jsonb *) DatumGetJsonbP(item);
+
+ JsonValueListClear(&scan->found);
+
+ MemoryContextResetOnly(scan->mcxt);
+
+ oldcxt = MemoryContextSwitchTo(scan->mcxt);
+
+ res = executeJsonPath(scan->path, scan->args,
+  GetJsonPathVar, CountJsonPathVars,
+  js, scan->errorOnError, &scan->found,
+  false /* FIXME */ );
+
+ MemoryContextSwitchTo(oldcxt);
+
+ if (jperIsError(res))
+ {
+ Assert(!scan->errorOnError);
+ JsonValueListClear(&scan->found); /* EMPTY ON ERROR case */
+ }
+
+ JsonTableRescan(scan);
+}

"FIXME".
set the last argument in executeJsonPath to true also works as expected.
also there is no test related to the "FIXME"
i am not 100% sure about the "FIXME".

see demo (after set the executeJsonPath's "useTz" argument to true).

create table ss(js jsonb);
INSERT into ss select '{"a": "2018-02-21 12:34:56 +10"}';
INSERT into ss select '{"b": "2018-02-21 12:34:56 "}';
PREPARE q2 as SELECT jt.*  FROM ss, JSON_TABLE(js, '$.a.datetime()'
COLUMNS ("int7" timestamptz PATH '$')) jt;
PREPARE qb as SELECT jt.*  FROM ss, JSON_TABLE(js, '$.b.datetime()'
COLUMNS ("tstz" timestamptz PATH '$')) jt;
PREPARE q3 as SELECT jt.*  FROM ss, JSON_TABLE(js, '$.a.datetime()'
COLUMNS ("ts" timestamp PATH '$')) jt;

begin;
set time zone +10;
EXECUTE q2;
set time zone -10;
EXECUTE q2;
rollback;

begin;
set time zone +10;
SELECT JSON_VALUE(js, '$.a' returning timestamptz) from ss;
set time zone -10;
SELECT JSON_VALUE(js, '$.a' returning timestamptz) from ss;
rollback;
---------------------------------------------------------------------
begin;
set time zone +10;
EXECUTE qb;
set time zone -10;
EXECUTE qb;
rollback;

begin;
set time zone +10;
SELECT JSON_VALUE(js, '$.b' returning timestamptz) from ss;
set time zone -10;
SELECT JSON_VALUE(js, '$.b' returning timestamptz) from ss;
rollback;
---------------------------------------------------------------------
begin;
set time zone +10;
EXECUTE q3;
set time zone -10;
EXECUTE q3;
rollback;

begin;
set time zone +10;
SELECT JSON_VALUE(js, '$.b' returning timestamp) from ss;
set time zone -10;
SELECT JSON_VALUE(js, '$.b' returning timestamp) from ss;
rollback;