The following bug has been logged online:

Bug reference:      5093
Logged by:          Mike Pomraning
Email address:      m...@pilcrow.madison.wi.us
PostgreSQL version: 8.4.1
Operating system:   Linux i686 2.6.18-128.7.1.el5
Description:        Prepared query gives different PGresult than exec'd
equivalent
Details: 

The following short program installs a RULE to SELECT two rows INSTEAD of
INSERTing into a VIEW.

When it PQexec's the insertion, I get a PGresult object with PQntuples == 2.
 However, when it PREPAREs/EXECUTEs the very same SQL, the PGresult has
PQntuples == 0.

My expectation is that the prepared statement would return the same PGresult
as its unprepared equivalent.

Am I using the API incorrectly, is my expectation amiss, is this a bug,
etc.?

/*
  $ ./a.out 'dbname=postgres password="mypass"'
  PGresult of exec() ('INSERT 0 0'): ntuples 2
  PGresult of execPrepd() ('INSERT 0 0'): ntuples 0
*/
#include <stdio.h>
#include <libpq-fe.h>

static char sql[] = "INSERT INTO v VALUES (1)";

int
main(int argc, char **argv) {
  PGconn *pg;
  PGresult *r;

  pg = PQconnectdb(argv[1] ? argv[1] : "");
  if (!pg || PQstatus(pg) == CONNECTION_BAD) {
    printf("connection failed\n");
    return 1;
  }

  PQexec(pg, "CREATE TEMPORARY VIEW v AS SELECT 1 WHERE 1=0");
  PQexec(pg, "CREATE RULE r AS ON INSERT TO v DO INSTEAD (SELECT 1 UNION
SELECT 2)");

  r = PQexec(pg, sql);
  printf("PGresult of exec() ('%s'): ntuples %d\n", PQcmdStatus(r),
PQntuples(r));

  PQprepare(pg, "pstmt", sql, 0, NULL);

  r = PQexecPrepared(pg, "pstmt", 0, NULL, NULL, NULL, 0);
  printf("PGresult of execPrepd() ('%s'): ntuples %d\n", PQcmdStatus(r),
PQntuples(r));

  PQexec(pg, "DROP RULE r");
  return 0;
}

-- 
Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-bugs

Reply via email to