Re: SQL/JSON: functions

Hello hackers,
On branch REL_15_STABLE the following query: SELECT
JSON_SERIALIZE('{"a":1}' RETURNING jsonb);

produces SIGSEGV for me:
#0 getJsonbOffset (index=39920251, jc=0x563cc5601d5c) at
jsonb_util.c:148
148 offset += JBE_OFFLENFLD(jc->children[i]);
(gdb) bt
#0 getJsonbOffset (index=39920251, jc=0x563cc5601d5c) at
jsonb_util.c:148
#1 JsonbIteratorNext (it=0x7ffdaff752c8, val=0x7ffdaff752d0,
skipNested=false) at jsonb_util.c:933
#2 0x0000563cc3e6934b in JsonbToCStringWorker (out=0x563cc55fe9a0,
in=<optimized out>, estimated_len=<optimized out>, indent=false) at
jsonb.c:496
#3 0x0000563cc3f35c68 in FunctionCall1Coll (arg1=<optimized out>,
collation=0, flinfo=0x563cc55fb8a0) at fmgr.c:1124
#4 OutputFunctionCall (flinfo=0x563cc55fb8a0, val=<optimized out>) at
fmgr.c:1561
#5 0x0000563cc3a8ebc5 in printtup (slot=0x563cc55faed8,
self=0x563cc5602758) at printtup.c:357
#6 0x0000563cc3c51917 in ExecutePlan (execute_once=<optimized out>,
dest=0x563cc5602758, direction=<optimized out>, numberTuples=0,
sendTuples=<optimized out>, operation=CMD_SELECT,
use_parallel_mode=<optimized out>, planstate=0x563cc55fabb8,
estate=0x563cc55fa980) at execMain.c:1667
#7 standard_ExecutorRun (queryDesc=0x563cc5553e60, direction=<optimized
out>, count=0, execute_once=<optimized out>) at execMain.c:363
#8 0x0000563cc3df741f in PortalRunSelect (portal=0x563cc55a3c40,
forward=<optimized out>, count=0, dest=<optimized out>) at pquery.c:924
#9 0x0000563cc3df8b81 in PortalRun (portal=portal(at)entry=0x563cc55a3c40,
count=count(at)entry=9223372036854775807, isTopLevel=isTopLevel(at)entry=true,
run_once=run_once(at)entry=true, dest=dest(at)entry=0x563cc5602758,
altdest=altdest(at)entry=0x563cc5602758,
qc=0x7ffdaff75650) at pquery.c:768
#10 0x0000563cc3df498d in exec_simple_query (query_string=0x563cc5531f00
"SELECT JSON_SERIALIZE('{\"a\":1}' RETURNING jsonb);") at
postgres.c:1250
#11 0x0000563cc3df651a in PostgresMain (dbname=<optimized out>,
username=<optimized out>) at postgres.c:4558
#12 0x0000563cc3d5b5e7 in BackendRun (port=<optimized out>,
port=<optimized out>) at postmaster.c:4504
#13 BackendStartup (port=<optimized out>) at postmaster.c:4232
#14 ServerLoop () at postmaster.c:1806
#15 0x0000563cc3d5c6da in PostmasterMain (argc=argc(at)entry=3,
argv=argv(at)entry=0x563cc552c6d0) at postmaster.c:1478
#16 0x0000563cc3a7b550 in main (argc=3, argv=0x563cc552c6d0) at
main.c:202

The first bad commit is 606948b058dc16bce494270eea577011a602810e

Andrew Dunstan писал 2022-04-01 03:25:
> On 3/31/22 15:54, Greg Stark wrote:
>> I see several commits referencing this entry. Can we mark it committed
>> or are there still chunks of commits to go?
>
>
>
> No code chunks left, only a documentation patch which should land
> tomorrow or Saturday.
>
>
> I am also planning on committing the JSON_TABLE patches before feature
> freeze (April 7). They depend on this set.
>
>
> cheers
>
>
> andrew
>
> --
> Andrew Dunstan
> EDB: https://www.enterprisedb.com