Problems with a C function, pg_uname(), and String concatenation.

From: Rafael Martinez <r(dot)m(dot)guerrero(at)usit(dot)uio(dot)no>
To: pgsql-general <pgsql-general(at)postgresql(dot)org>
Subject: Problems with a C function, pg_uname(), and String concatenation.
Date: 2008-07-01 08:45:07
Message-ID: 4869EE93.2000507@usit.uio.no
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Hello

We have a function in C which is accessed via a view and which produces
a strange result when used together with || (String concatenation).

I can not find the problem. Any C/postgres guru with any idea of how to
fix it?

Here is the function and the result when used with ||:

-------------------------------------------
PG_UNAME()
-------------------------------------------
#include "postgres.h"
#include <string.h>
#include "fmgr.h"
#include <sys/utsname.h>

#ifdef PG_MODULE_MAGIC
PG_MODULE_MAGIC;
#endif

PG_FUNCTION_INFO_V1(pg_uname);

Datum
pg_uname(PG_FUNCTION_ARGS)
{
text *argument = PG_GETARG_TEXT_P(0);
size_t argumentlen = VARSIZE(argument)-VARHDRSZ;

text *result = (text *) palloc(256);
char *option = (char *) palloc(argumentlen+1);

char sysname[] = "sysname";
char nodename[] = "nodename";
char release[] = "release";
char version[] = "version";
char machine[] = "machine";
char null[] = "null";

struct utsname uname_pointer;
uname(&uname_pointer);

VARATT_SIZEP(result) = 256;

memcpy(option,VARDATA(argument),argumentlen);
option[argumentlen] = '\0';

if (strcmp(option,sysname) == 0){

memcpy(VARDATA(result),uname_pointer.sysname,sizeof(uname_pointer.sysname));
}
else if (strcmp(option,nodename) == 0){

memcpy(VARDATA(result),uname_pointer.nodename,sizeof(uname_pointer.nodename));
}
else if (strcmp(option,release) == 0){

memcpy(VARDATA(result),uname_pointer.release,sizeof(uname_pointer.release));
}
else if (strcmp(option,version) == 0){

memcpy(VARDATA(result),uname_pointer.version,sizeof(uname_pointer.version));
}
else if (strcmp(option,machine) == 0){

memcpy(VARDATA(result),uname_pointer.machine,sizeof(uname_pointer.machine));
}
else{
memcpy(VARDATA(result),null,sizeof(null));
}

pfree(option);
PG_RETURN_TEXT_P(result);
}

-------------------------------------------

The view we use is this one:
-------------------------------------------
SELECT pg_uname('sysname'::text) AS sysname,
pg_uname('nodename'::text) AS nodename,
pg_uname('release'::text) AS kernel,
pg_uname('version'::text) AS version,
pg_uname('machine'::text) AS machine,
substr(version(), 11, 7) AS pgversion,
to_char(pg_postmaster_start_time(),'YYYY-MM-DD
HH24:MM:SS'::text) AS pg_start,
age(now(),pg_postmaster_start_time()) AS pg_uptime;
-------------------------------------------

A normal output from this view looks like this:

pgadmin=# SELECT * from sysinfo;

-[ RECORD 1 ]----------------------------------
sysname | Linux
nodename | dbpg-rt.uio.no
kernel | 2.6.9-67.0.4.ELsmp
version | #1 SMP Fri Jan 18 05:00:00 EST 2008
machine | x86_64
pgversion | 8.2.6
pg_start | 2008-06-10 10:06:28
pg_uptime | 20 days 23:59:24.971497

But, when used with ||, everything comming after attributes delivered by
pg_uname() disappears:

SELECT '***' || sysname || '***' as example from sysinfo ;

example
----------
***Linux
(1 row)

Any ideas?
Thanks in advance for your time.

PS.- This is the section og the make file used to compile this function:
-----------------------------------------------------------------------
SERVER_INCLUDES += -I $(shell /usr/local/bin/pg_config --includedir)
SERVER_INCLUDES += -I $(shell /usr/local/bin/pg_config --includedir-server)

CFLAGS = $(SERVER_INCLUDES)
CC = gcc

pg_uname: pg_uname.c
$(CC) $(CFLAGS) -fpic -c $<
$(CC) $(CFLAGS) -shared -o $(basename $<).so $(basename $<).o
-----------------------------------------------------------------------

regards
--
Rafael Martinez, <r(dot)m(dot)guerrero(at)usit(dot)uio(dot)no>
Center for Information Technology Services
University of Oslo, Norway

PGP Public Key: http://folk.uio.no/rafael/

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Richard Huxton 2008-07-01 08:58:26 Re: Query Fails
Previous Message Jamie Deppeler 2008-07-01 08:37:07 Need some help