使用C编写的动态链接库为PostgreSQL数据库添加自定义函数

不得不说，在这点上，PostgreSQL功能非常强大，不亚于任何家大型商业数据库。要添加一些自定义函数，不是那么复杂。

下边的示例会添加如下函数：test(), concat_text(string, string)  还有另外两个，不作举例，您可以自行尝试

1. 代码如下：

/* $PostgreSQL: pgsql/src/tutorial/funcs_new.c,v 1.13 2007/02/27 23:48:10 tgl Exp $ */

/******************************************************************************
  These are user-defined functions that can be bound to a Postgres backend
  and called by Postgres to execute SQL functions of the same name.

  The calling format for these functions is defined by the CREATE FUNCTION
  SQL statement that binds them to the backend.

  NOTE: this file shows examples of "new style" function call conventions.
  See funcs.c for examples of "old style".
*****************************************************************************/

#include "postgres.h"			/* general Postgres declarations */

#include "executor/executor.h"	/* for GetAttributeByName() */
#include "utils/geo_decls.h"	/* for point type */

#include "executor/spi.h"

PG_MODULE_MAGIC;

/* These prototypes just prevent possible warnings from gcc. */
//
//Datum		add_one(PG_FUNCTION_ARGS);
//Datum		add_one_float8(PG_FUNCTION_ARGS);
//Datum		makepoint(PG_FUNCTION_ARGS);
//Datum		copytext(PG_FUNCTION_ARGS);
//Datum		concat_text(PG_FUNCTION_ARGS);
//Datum		c_overpaid(PG_FUNCTION_ARGS);


/* By Value */

PG_FUNCTION_INFO_V1(add_one);

__declspec (dllexport) Datum
add_one(PG_FUNCTION_ARGS)
{
	int32		arg = PG_GETARG_INT32(0);

	PG_RETURN_INT32(arg + 1);
}

/* By Reference, Fixed Length */

PG_FUNCTION_INFO_V1(add_one_float8);

__declspec (dllexport) Datum
add_one_float8(PG_FUNCTION_ARGS)
{
	/* The macros for FLOAT8 hide its pass-by-reference nature */
	float8		arg = PG_GETARG_FLOAT8(0);

	PG_RETURN_FLOAT8(arg + 1.0);
}

PG_FUNCTION_INFO_V1(makepoint);

__declspec (dllexport) Datum
makepoint(PG_FUNCTION_ARGS)
{
	Point	   *pointx = PG_GETARG_POINT_P(0);
	Point	   *pointy = PG_GETARG_POINT_P(1);
	Point	   *new_point = (Point *) palloc(sizeof(Point));

	new_point->x = pointx->x;
	new_point->y = pointy->y;

	PG_RETURN_POINT_P(new_point);
}

/* By Reference, Variable Length */

PG_FUNCTION_INFO_V1(copytext);

__declspec (dllexport) Datum
copytext(PG_FUNCTION_ARGS)
{
	text	   *t = PG_GETARG_TEXT_P(0);

	/*
	 * VARSIZE is the total size of the struct in bytes.
	 */
	text	   *new_t = (text *) palloc(VARSIZE(t));

	SET_VARSIZE(new_t, VARSIZE(t));

	/*
	 * VARDATA is a pointer to the data region of the struct.
	 */
	memcpy((void *) VARDATA(new_t),		/* destination */
		   (void *) VARDATA(t), /* source */
		   VARSIZE(t) - VARHDRSZ);		/* how many bytes */
	PG_RETURN_TEXT_P(new_t);
}

PG_FUNCTION_INFO_V1(concat_text);

__declspec (dllexport) Datum
concat_text(PG_FUNCTION_ARGS)
{
	text	   *arg1 = PG_GETARG_TEXT_P(0);
	text	   *arg2 = PG_GETARG_TEXT_P(1);
	int32		arg1_size = VARSIZE(arg1) - VARHDRSZ;
	int32		arg2_size = VARSIZE(arg2) - VARHDRSZ;
	int32		new_text_size = arg1_size + arg2_size + VARHDRSZ;
	text	   *new_text = (text *) palloc(new_text_size);

	SET_VARSIZE(new_text, new_text_size);
	memcpy(VARDATA(new_text), VARDATA(arg1), arg1_size);
	memcpy(VARDATA(new_text) + arg1_size, VARDATA(arg2), arg2_size);
	PG_RETURN_TEXT_P(new_text);
}

/* Composite types */

PG_FUNCTION_INFO_V1(c_overpaid);

__declspec (dllexport) Datum
c_overpaid(PG_FUNCTION_ARGS)
{
	HeapTupleHeader t = PG_GETARG_HEAPTUPLEHEADER(0);
	int32		limit = PG_GETARG_INT32(1);
	bool		isnull;
	int32		salary;

	salary = DatumGetInt32(GetAttributeByName(t, "salary", &isnull));
	if (isnull)
		PG_RETURN_BOOL(false);

	/*
	 * Alternatively, we might prefer to do PG_RETURN_NULL() for null salary
	 */

	PG_RETURN_BOOL(salary > limit);
}

PG_FUNCTION_INFO_V1(test);
__declspec (dllexport)  Datum test(PG_FUNCTION_ARGS)
{
       char *command;
       int ret;
       int proc;

       command = "SELECT 1";

       elog(INFO,"1");

       SPI_connect();

       elog(INFO,"2");

       ret = SPI_exec(command, 1);

       elog(INFO,"3");

       proc = SPI_processed;

       elog(INFO,"4");
       if (ret > 0 && SPI_tuptable != NULL)
       {
           TupleDesc tupdesc = SPI_tuptable->tupdesc;
           SPITupleTable *tuptable = SPI_tuptable;
           char buf[8192];
           int i, j;

           elog(INFO,"5");
           for (j = 0; j < proc; j++)
           {
               HeapTuple tuple = tuptable->vals[j];

               elog(INFO,"6");
               for (i = 1, buf[0] = 0; i <= tupdesc->natts; i++)
               {
                       snprintf(buf + strlen (buf), sizeof(buf) - strlen(buf), " %s%s",
                                SPI_getvalue(tuple, tupdesc, i),
                                (i == tupdesc->natts) ? " " : " |");

                       elog(INFO,"7");
               }
               elog(INFO, "EXECQ: %s", buf);
           }
       }

       SPI_finish();
       // pfree(command);

}

2. 编译过程

要注意的是编译过程，需要添加头文件路径，lib路径，bin路径，以完成编译，最后得到func1.dll，以本例为例，过程并不复杂。在linux平台，相信更简单。

以windows为例，您可以先从http://download.csdn.net/detail/iihero/2979577 下载整个示例工程，应该是vs2005的vc++工程，添加如下include路径，最前头带PG(9.0或以上)的安装路径：

include\server
include\server\port\win32
include\server\port\win32_msvc
include\

最后得到func1.dll，将其复制到${PG_HOME}\bin下边。

3. 验证创建的函数

create function mytest() RETURNs VARCHAR(256)
AS 'd:/pg921/bin/func1', 'test'
LANGUAGE C STRICT;

说明： 添加SQL函数mytest, 不带参数，它的实现是func1.dll中的test，含义非常明确。  

iihero=# create function mytest() RETURNs VARCHAR(256)
iihero-# AS 'd:/pg921/bin/func1', 'test'
iihero-# LANGUAGE C STRICT;
CREATE FUNCTION
iihero=# select mytest();
INFO:  1
INFO:  2
INFO:  3
INFO:  4
INFO:  5
INFO:  6
INFO:  7
INFO:  EXECQ:  1
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: Succeeded.


create function myconcat(varchar(256), varchar(256)) RETURNs VARCHAR(512)
AS 'd:/pg921/bin/func1', 'concat_text'
LANGUAGE C STRICT;

说明： 添加函数myconcat, 带两个字符串参数，它的实现是func1.dll中的concat_text，含义非常明确。

iihero=# create function myconcat(varchar(256), varchar(256)) RETURNs VARCHAR(512)
iihero-# AS 'd:/pg921/bin/func1', 'concat_text'
iihero-# LANGUAGE C STRICT;
CREATE FUNCTION
iihero=# select myconcat('abc###, ', 'bcd1234');
    myconcat
-----------------
 abc###, bcd1234
(1 row)


iihero=#

其实，在其它商业数据库里，至少Sybase ASE/ASA, Oracle是支持这种C函数扩展的，就是编写起来，可能更麻烦一些。