Horton

Re: pgsql: Add more SQL/JSON constructor functions

jian he <jian.universality@gmail.com>

View thread View on postgr.es
From: jian he <jian.universality@gmail.com>
To: Amit Langote <amitlangote09@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Peter Eisentraut <peter@eisentraut.org>, Alvaro Herrera <alvherre@alvh.no-ip.org>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2024-06-28T06:13:53Z
Lists: pgsql-hackers 
On Thu, Jun 27, 2024 at 7:48 PM Amit Langote <amitlangote09@gmail.com> wrote:
> >
> > I've attempted that in the attached 0001, which removes
> > JsonExpr.coercion_expr and a bunch of code around it.
> >
> > 0002 is now the original patch minus the changes to make
> > JSON_EXISTS(), JSON_QUERY(), and JSON_VALUE() behave as we would like,
> > because the changes in 0001 covers them. The changes for JsonBehavior
> > expression coercion as they were in the last version of the patch are
> > still needed, but I decided to move those into 0001 so that the
> > changes for query functions are all in 0001 and those for constructors
> > in 0002.  It would be nice to get rid of that coerce_to_target_type()
> > call to coerce the "behavior expression" to RETURNING type, but I'm
> > leaving that as a task for another day.
>
> Updated 0001 to remove outdated references, remove some more unnecessary code.
>

i found some remaining references of "coercion_expr" should be removed.

src/include/nodes/primnodes.h
/* JsonExpr's collation, if coercion_expr is NULL. */


src/include/nodes/execnodes.h
/*
* Address of the step to coerce the result value of jsonpath evaluation
* to the RETURNING type using JsonExpr.coercion_expr.  -1 if no coercion
* is necessary or if either JsonExpr.use_io_coercion or
* JsonExpr.use_json_coercion is true.
*/
int jump_eval_coercion;

src/backend/jit/llvm/llvmjit_expr.c
/* coercion_expr code */
LLVMPositionBuilderAtEnd(b, b_coercion);
if (jsestate->jump_eval_coercion >= 0)
LLVMBuildBr(b, opblocks[jsestate->jump_eval_coercion]);
else
LLVMBuildUnreachable(b);


src/backend/executor/execExprInterp.c
/*
 * Checks if an error occurred either when evaluating JsonExpr.coercion_expr or
 * in ExecEvalJsonCoercion().  If so, this sets JsonExprState.error to trigger
 * the ON ERROR handling steps.
 */
void
ExecEvalJsonCoercionFinish(ExprState *state, ExprEvalStep *op)
{
}

if (jbv == NULL)
{
/* Will be coerced with coercion_expr, if any. */
*op->resvalue = (Datum) 0;
*op->resnull = true;
}


src/backend/executor/execExpr.c
/*
* Jump to coerce the NULL using coercion_expr if present.  Coercing NULL
* is only interesting when the RETURNING type is a domain whose
* constraints must be checked.  jsexpr->coercion_expr containing a
* CoerceToDomain node must have been set in that case.
*/

/*
* Jump to coerce the NULL using coercion_expr if present.  Coercing NULL
* is only interesting when the RETURNING type is a domain whose
* constraints must be checked.  jsexpr->coercion_expr containing a
* CoerceToDomain node must have been set in that case.
*/

Commits

  1. SQL/JSON: Avoid initializing unnecessary ON ERROR / ON EMPTY steps

  2. SQL/JSON: Fix default ON ERROR behavior for JSON_TABLE

  3. SQL/JSON: Fix JSON_TABLE() column deparsing

  4. Update comment about ExprState.escontext

  5. SQL/JSON: Fix casting for integer EXISTS columns in JSON_TABLE

  6. SQL/JSON: Some fixes to JsonBehavior expression casting

  7. SQL/JSON: Remove useless code in ExecInitJsonExpr()

  8. SQL/JSON: Respect OMIT QUOTES when RETURNING domains over jsonb

  9. SQL/JSON: Improve error-handling of JsonBehavior expressions

  10. SQL/JSON: Fix error-handling of some JsonBehavior expressions

  11. SQL/JSON: Rethink c2d93c3802b

  12. SQL/JSON: Always coerce JsonExpr result at runtime

  13. SQL/JSON: Fix coercion of constructor outputs to types with typmod