工作关系 同时使用SQL SERVER 与Postgresql 数据库，

            clientChar = txtClientChar.Text.Trim();
            sql = "select client_id,client_code,name,client_level_id,client_custom_options_ids, area_id,location ";
            sql += " from client where client_code = @pClient_Code or name like @pName";
            NpgsqlParameter[] sqlPara = new NpgsqlParameter[2];
            //pgsql 的参数与SQL SERVER 的格式不一样
            sqlPara[0] = new NpgsqlParameter("@pClient_Code", NpgsqlTypes.NpgsqlDbType.Text,20);
            sqlPara[0].Value = clientChar;
            sqlPara[1] = new NpgsqlParameter("@pName", NpgsqlTypes.NpgsqlDbType.Text,100);
            sqlPara[1].Value = '%' + clientChar + '%';
            DataTable dt = DBHelperPg.ExecuteDataTable(sql, sqlPara);
            dataGridView1.DataSource = dt;

附：DBHelperpg.cs

 1  class DBHelperPg
 2     {
 3 
 4         public static string ConnectionString = ConfigurationManager.AppSettings["postgre"].ToString();
 5 
 6 
 7         /// <summary>  
 8         /// 执行SQL语句  
 9         /// </summary>  
10         /// <param name="sql">SQL</param>  
11         /// <returns>成功返回大于0的数字</returns>  
12         public static int ExecuteSQL(string sql)
13         {
14             int num2 = -1;
15             using (NpgsqlConnection connection = new NpgsqlConnection(ConnectionString))
16             {
17                 using (NpgsqlCommand command = new NpgsqlCommand(sql, connection))
18                 {
19                     try
20                     {
21                         connection.Open();
22                         num2 = command.ExecuteNonQuery();
23                     }
24                     catch (NpgsqlException exception)
25                     {
26                         throw new Exception(exception.Message);
27                     }
28                     finally
29                     {
30                         connection.Close();
31                     }
32                 }
33             }
34             return num2;
35         }
36 
37         //带参数的执行查询，不返回结果，返回影响行数
38         //执行SQL语句并返回受影响的行数
39         public static int ExecuteNonQuery(string sql, params NpgsqlParameter[] parameters)
40         {
41             using (NpgsqlConnection conn = new NpgsqlConnection(ConnectionString))
42             {
43                 conn.Open();
44                 using (NpgsqlCommand cmd = conn.CreateCommand())
45                 {
46                     cmd.CommandText = sql;
47                     //foreach (SqlParameter param in parameters)
48                     //{
49                     //    cmd.Parameters.Add(param);
50                     //}
51                     cmd.Parameters.AddRange(parameters);
52                     return cmd.ExecuteNonQuery();
53                 }
54             }
55         }
56 
57         //执行查询，并返回查询所返回的结果集中第一行的第一列。忽略额外的列或行。
58         public static object ExecuteScalar(string sql, params NpgsqlParameter[] parameters)
59         {
60             using (NpgsqlConnection conn = new NpgsqlConnection(ConnectionString))
61             {
62                 conn.Open();
63                 using (NpgsqlCommand cmd = conn.CreateCommand())
64                 {
65                     cmd.CommandText = sql;
66                     cmd.Parameters.AddRange(parameters);
67                     return cmd.ExecuteScalar();
68 
69                 }
70             }
71         }
72 
73 
74 
75 
76         //查询并返回结果集DataTable,一般只用来执行查询结果比较少的sql。
77         public static DataTable ExecuteDataTable(string sql, params NpgsqlParameter[] parameters)
78         {
79             using (NpgsqlConnection conn = new NpgsqlConnection(ConnectionString))
80             {
81                 conn.Open();
82                 using (NpgsqlCommand cmd = conn.CreateCommand())
83                 {
84                     cmd.CommandText = sql;
85                     cmd.Parameters.AddRange(parameters);
86 
87                     NpgsqlDataAdapter adapter = new NpgsqlDataAdapter(cmd);
88                     DataSet dataset = new DataSet();
89                     adapter.Fill(dataset);
90                     return dataset.Tables[0];
91                 }
92             }
93 
94             //查询较大的数据用 DateRead()，但应尽可能用分页数据，仍然用datatable更好。
95         }
96     }