当前位置：首页 > news >正文

unixODBC编程（七）数组查询

news 2025/5/1 3:35:22

数组查询与单条查询编程的步骤差不多，只是在执行语句前要设置4个语句句柄属性，在绑定输出变量时使用变量数组，先看一看需要哪四个句柄属性。

第一个属性告诉语句使用列数组绑定，属性名称为SQL_ATTR_ROW_BIND_TYPE，属性值为SQL_BIND_BY_COLUMN，例如：

SQLSetStmtAttr(stmth, SQL_ATTR_ROW_BIND_TYPE, SQL_BIND_BY_COLUMN, 0);

第二个属性告诉语句绑定的数组大小，就是绑定的行数，属性名称为SQL_ATTR_ROW_ARRAY_SIZE，属性值为10，例如：

SQLSetStmtAttr(stmth, SQL_ATTR_ROW_ARRAY_SIZE, (SQLPOINTER)10, 0);

第三个属性告诉语句返回每一行的状态，属性名称为SQL_ATTR_ROW_STATUS_PTR，属性值为一个SQLUSMALLINT类型的数组，比如叫status[10]，例如：

SQLSetStmtAttr(stmth, SQL_ATTR_ROW_STATUS_PTR, (SQLPOINTER)status, 0);

第四个属性告诉语句返回每次从结果集中取回的数据条数，属性名称为SQL_ATTR_ROWS_FETCHED_PTR，属性值为一个SQLULEN类型的整数，不是数组，比如叫做num_fetched，例如：

SQLSetStmtAttr(stmth, SQL_ATTR_ROWS_FETCHED_PTR, (SQLPOINTER)&num_fetched, 0);

在使用SQLBindCol()函数绑定输出变量时，TargetValuePtr参数使用数组绑定，StrLen_or_IndPtr参数也使用数组绑定，例如绑定一个整数数组id[10]，返回长度和指示变量数组len_id1[10]，调用函数如下：

SQLBindCol(stmth, 1, SQL_C_ULONG, (SQLPOINTER)id, 0, len_ind1);

下面看一个完整的例子，还是从前面创建的表test_tab1中查询数据，使用数组操作。

#include "stdio.h"
#include "stdlib.h"
#include "string.h"
#include "sql.h"
#include "sqlext.h"
#include "sqltypes.h"SQLHANDLE       envh;           /* env handle */
SQLHANDLE       dbch;           /* connect handle */
SQLHANDLE       stmth;          /* statement handle */int main(int argc, char *argv[])
{int                 i;int                 conn = 0;SQLRETURN           rc;SQLULEN             num_fetched;SQLUSMALLINT        status[10];SQLLEN              len_ind1[10];SQLLEN              len_ind2[10];SQLLEN              len_ind3[10];SQLLEN              len_ind4[10];SQLLEN              len_ind5[10];SQLINTEGER          id[10];char                dsn_str[32];char                usrname[32];char                passwd[32];char                sqltxt[128];char                f1[10][32];char                f2[10][32];char                f3[10][32];char                f4[10][32];if (argc < 3) {fprintf(stderr, "usage: %s dsn username password\n", argv[0]);return (-1);}strncpy(dsn_str, argv[1], 32);dsn_str[31] = '\0';strncpy(usrname, argv[2], 32);usrname[31] = '\0';strncpy(passwd, argv[3], 32);passwd[31] = '\0';rc = SQLAllocHandle(SQL_HANDLE_ENV, SQL_NULL_HANDLE, &envh);if (rc != SQL_SUCCESS) {fprintf(stderr, "Allocate environment handle error.\n");return (-1);}rc = SQLSetEnvAttr(envh, SQL_ATTR_ODBC_VERSION, (void *)SQL_OV_ODBC3, 0);if (rc != SQL_SUCCESS) {fprintf(stderr, "Set ODBC version error.\n");goto free_exit;}rc = SQLAllocHandle(SQL_HANDLE_DBC, envh, &dbch);if (rc != SQL_SUCCESS) {fprintf(stderr, "Allocate DB connection handle error.\n");goto free_exit;}rc = SQLSetConnectAttr(dbch, SQL_ATTR_LOGIN_TIMEOUT, (SQLPOINTER)10, 0);if (rc != SQL_SUCCESS) {fprintf(stderr, "Set connection timeout value error.\n");goto free_exit;}rc = SQLConnect(dbch, (SQLCHAR *)dsn_str, SQL_NTS, (SQLCHAR *)usrname, SQL_NTS,(SQLCHAR *)passwd, SQL_NTS);if (rc != SQL_SUCCESS) {fprintf(stderr, "Connect to DB error.\n");goto free_exit;}conn = 1;fprintf(stdout, "connect DB ok ......\n");rc = SQLAllocHandle(SQL_HANDLE_STMT, dbch, &stmth);if (rc != SQL_SUCCESS) {fprintf(stderr, "Allocate statment handle error.\n");goto free_exit;}sprintf(sqltxt, "select id, f1, f2, f3, f4 from test_tab1");rc = SQLPrepare(stmth, (SQLCHAR *)sqltxt, SQL_NTS);if (rc != SQL_SUCCESS) {fprintf(stderr, "Prepare statment error.\n");goto free_exit;}/* 使用列数组绑定 */rc = SQLSetStmtAttr(stmth, SQL_ATTR_ROW_BIND_TYPE, SQL_BIND_BY_COLUMN, 0);if (rc != SQL_SUCCESS) {fprintf(stderr, "Set statment attribute error.\n");goto free_exit;}/* 设置数组大小为10，每次取回10条数据 */rc = SQLSetStmtAttr(stmth, SQL_ATTR_ROW_ARRAY_SIZE, (SQLPOINTER)10, 0);if (rc != SQL_SUCCESS) {fprintf(stderr, "Set statement attribute error.\n");goto free_exit;}/* 设置返回每条数据的状态数组 */rc = SQLSetStmtAttr(stmth, SQL_ATTR_ROW_STATUS_PTR, (SQLPOINTER)status, 0);if (rc != SQL_SUCCESS) {fprintf(stderr, "Set statement attribute error.\n");goto free_exit;}/* 设置返回每次取回数据的条数 */rc = SQLSetStmtAttr(stmth, SQL_ATTR_ROWS_FETCHED_PTR, (SQLPOINTER)&num_fetched, 0);if (rc != SQL_SUCCESS) {fprintf(stderr, "Set statement attribute error.\n");goto free_exit;}/* 执行语句 */rc = SQLExecute(stmth);if (rc != SQL_SUCCESS) {fprintf(stderr, "Execute statment error.\n");goto free_exit;}/* 绑定第一列，id使用数组，len_ind1也使用数组 */rc = SQLBindCol(stmth, 1, SQL_C_ULONG, (SQLPOINTER)id, 0, len_ind1);if (rc != SQL_SUCCESS) {fprintf(stderr, "Bind column 1 error.\n");goto free_exit;}rc = SQLBindCol(stmth, 2, SQL_C_CHAR, (SQLPOINTER)f1, 32, len_ind2);if (rc != SQL_SUCCESS) {fprintf(stderr, "Bind column 2 error.\n");goto free_exit;}rc = SQLBindCol(stmth, 3, SQL_C_CHAR, (SQLPOINTER)f2, 32, len_ind3);if (rc != SQL_SUCCESS) {fprintf(stderr, "Bind column 3 error.\n");goto free_exit;}rc = SQLBindCol(stmth, 4, SQL_C_CHAR, (SQLPOINTER)f3, 32, len_ind4);if (rc != SQL_SUCCESS) {fprintf(stderr, "Bind column 4 error.\n");goto free_exit;}rc = SQLBindCol(stmth, 5, SQL_C_CHAR, (SQLPOINTER)f4, 32, len_ind5);if (rc != SQL_SUCCESS) {fprintf(stderr, "Bind column 5 error.\n");goto free_exit;}while (1) {rc = SQLFetch(stmth);if (rc == SQL_NO_DATA) {fprintf(stderr, "No data in result set, break.\n");break;} else if (rc == SQL_ERROR) {fprintf(stderr, "Fetch data error.\n");goto free_exit;}fprintf(stdout, "%lu rows fetched.\n", num_fetched);for (i=0; i<num_fetched; i++) {fprintf(stdout, "id=%d, f1=%s, f2=%s, f3=%s, f4=%s\n",id[i], f1[i], f2[i], f3[i], f4[i]);}}SQLFreeHandle(SQL_HANDLE_STMT, stmth);SQLDisconnect(dbch);SQLFreeHandle(SQL_HANDLE_DBC, dbch);SQLFreeHandle(SQL_HANDLE_ENV, envh);return (0);free_exit:if (stmth != NULL) {SQLFreeHandle(SQL_HANDLE_STMT, stmth);}if (conn) {SQLDisconnect(dbch);}if (dbch != NULL) {SQLFreeHandle(SQL_HANDLE_DBC, dbch);}if (envh != NULL) {SQLFreeHandle(SQL_HANDLE_ENV, envh);}return (-1);
}

访问www.tomcoding.com网站，可以下载Oracle内部数据结构的文档，还有DUL，exp/imp，logminer及ASM工具的源代码。

查看全文

http://www.mrgr.cn/news/39406.html