Re: remaining sql/json patches

05.04.2024 10:09, Amit Langote wrote:
> Seems like it might be a pre-existing issue, because I can also
> reproduce the crash with:
>
> SELECT * FROM COALESCE(row(1)) AS (a int, b int);
> server closed the connection unexpectedly
> This probably means the server terminated abnormally
> before or while processing the request.
> The connection to the server was lost. Attempting reset: Failed.
> !>
>
> Backtrace:
>
> #0 __pthread_kill_implementation (threadid=281472845250592,
> signo=signo(at)entry=6, no_tid=no_tid(at)entry=0) at pthread_kill.c:44
> #1 0x0000ffff806c4334 in __pthread_kill_internal (signo=6,
> threadid=<optimized out>) at pthread_kill.c:78
> #2 0x0000ffff8067c73c in __GI_raise (sig=sig(at)entry=6) at
> ../sysdeps/posix/raise.c:26
> #3 0x0000ffff80669034 in __GI_abort () at abort.c:79
> #4 0x0000000000ad9d4c in ExceptionalCondition (conditionName=0xcbb368
> "!(tupdesc->natts >= colcount)", errorType=0xcbb278 "FailedAssertion",
> fileName=0xcbb2c8 "nodeFunctionscan.c",
> lineNumber=379) at assert.c:54

That's strange, because I get the error (on master, 6f132ed69).
With backtrace_functions = 'tupledesc_match', I see
2024-04-05 10:48:27.827 MSK client backend[2898632] regress ERROR: function return row and query-specified return row do
not match
2024-04-05 10:48:27.827 MSK client backend[2898632] regress DETAIL: Returned row contains 1 attribute, but query expects 2.
2024-04-05 10:48:27.827 MSK client backend[2898632] regress BACKTRACE:
tupledesc_match at execSRF.c:948:3
ExecMakeTableFunctionResult at execSRF.c:427:13
FunctionNext at nodeFunctionscan.c:94:5
ExecScanFetch at execScan.c:131:10
ExecScan at execScan.c:180:10
ExecFunctionScan at nodeFunctionscan.c:272:1
ExecProcNodeFirst at execProcnode.c:465:1
ExecProcNode at executor.h:274:9
 (inlined by) ExecutePlan at execMain.c:1646:10
standard_ExecutorRun at execMain.c:363:3
ExecutorRun at execMain.c:305:1
PortalRunSelect at pquery.c:926:26
PortalRun at pquery.c:775:8
exec_simple_query at postgres.c:1282:3
PostgresMain at postgres.c:4684:27
BackendMain at backend_startup.c:57:2
pgarch_die at pgarch.c:847:1
BackendStartup at postmaster.c:3593:8
ServerLoop at postmaster.c:1674:6
main at main.c:184:3
        /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0x80) [0x7f37127f0e40]
2024-04-05 10:48:27.827 MSK client backend[2898632] regress STATEMENT:  SELECT * FROM COALESCE(row(1)) AS (a int, b int);

That's why I had attributed the failure to JSON_TABLE().

Though SELECT * FROM generate_series(1, 1), COALESCE(row(1)) AS (a int, b int);
really triggers the assert too.
Sorry for the noise...

Best regards,
Alexander