Get reusable DB connection

本文介绍如何在SoapUI中创建可复用的数据库连接,以提高测试效率并减少资源消耗。通过设置测试案例属性及使用Groovy脚本,可以在测试开始时建立数据库连接,并在测试结束后关闭。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

Refer to:
1. http://onebyteatatime.wordpress.com/2009/02/17/reusable-sql-connection-in-soapui/

2. http://groovyinsoapui.wordpress.com/2008/09/08/getting-db-connection-with-groovy/

3. http://www.soapui.org/userguide/functional/groovystep.html#soapUI_GroovyUtils

4.http://groovy.codehaus.org/Database+features

....

Reusable SQL Connection in soapUI

No matter how modern the systems get, we always often have some level of interaction with databases whether for reporting purposes or for recording transactions. In typical web services testing, one would imagine testing xml response to a given xml payload. What if, you need to connect to a database and say, verify, if a particular record was inserted by firing a select query? OR if, you need to create a log of your testing right inside the database? If not you may need to fire a select query after each SOAP Request, right in the Groovy Assertion Step of that request.

Now, quick and dirty way would be to create a SQL connection each time you need a connection and close it once you are done with it. Creating and closing a resource connection is very expensive. Hence, this could prove very costly when you are testing thousands of SOAP Requests. Imagine for each request you fire if there is a connection created whether in Groovy Assertion step or otherwise! I shall try to go over a simple mechanism to create a common reusable connection right at the beginning of your soapUI TestCase and then recycle it once you don’t need it.

The example I am trying to illustrate shall be useful when you have a single soapUI test case that, say, is iterating over number of rows from a datasource, carrying out some number of test steps and at each test step, before the step or after that step you need to connect to the database and carry out some operation.

To begin for any RDBMS you need four things: Driver, URL, Username and Password.
1. You can store these as TestCase properties, say, “dbUrl”, “dbDriverClassName”, “dbUser”, “dbPassword”

    For example: dbUrl:jdbc:microsoft:sqlserver://mssqlops1.prn-corp.com

                           dbDriverClassName:  com.microsoft.jdbc.sqlserver.SQLServerDriver

                           .......
2. Each soapUI test case has two scripts: Setup Script and TearDown Script. We shall make use of these two scripts in order to illustrate our example. Inside Setup Script we shall obtain handle to our common connection as shown below:

  1.   
  2. //In a Setup Script   
  3. import groovy.sql.Sql   
  4.   
  5. //try to create connection to database, if available. load this connection on context   
  6. //if not, log error and continue   
  7. //In order for this to work, you need to have jdbc driver jar file in $SOAPUI_HOME/bin/ext folder   
  8. def url = context.expand( '${#TestCase#dbUrl}' )   
  9. def driver = context.expand( '${#TestCase#dbDriverClassName}' )   
  10. def user = context.expand( '${#TestCase#dbUser}' )   
  11. def password = context.expand( '${#TestCase#dbPassword}' )   
  12.   
  13. if ( (null != url) && (null != driver) && (null != user) && (null != password) )   
  14. {   
  15.   try {   
  16.     connection = Sql.newInstance(url, user, password, driver)   
  17.     context.setProperty("dbConn", connection)   
  18.   } catch (Exception e) {   
  19.     log.error "Could not establish connection to the database."  
  20.   }   
  21. }  
//In a Setup Script
import groovy.sql.Sql

//try to create connection to database, if available. load this connection on context
//if not, log error and continue
//In order for this to work, you need to have jdbc driver jar file in $SOAPUI_HOME/bin/ext folder
def url = context.expand( '${#TestCase#dbUrl}' )
def driver = context.expand( '${#TestCase#dbDriverClassName}' )
def user = context.expand( '${#TestCase#dbUser}' )
def password = context.expand( '${#TestCase#dbPassword}' )

if ( (null != url) && (null != driver) && (null != user) && (null != password) )
{
  try {
    connection = Sql.newInstance(url, user, password, driver)
    context.setProperty("dbConn", connection)
  } catch (Exception e) {
    log.error "Could not establish connection to the database."
  }
}

3. Now wherever we need access to this connection in any of our test steps: whether in Groovy Assertion or in Groovy Script, we simply check for existence of this property on the soapUI context, for example, say following Groovy Assertion verifies if a customer account was created in the database right after SOAP Request CreateCustomer.

  1. //In a Groovy Assertion Step of a SOAP Request   
  2. def MSG_CUSTOMER_NOT_FOUND = "Customer not found!"  
  3. // Obtain customer number from response of a SOAP Request that creates a customer   
  4. def customerNumber = <... groovy code retrieving customer number from response here ...>    
  5.   
  6. //Check if connection to database is available   
  7. if (context.dbConn)   
  8. {   
  9.   //connection to the database   
  10.   def sql = context.dbConn   
  11.   
  12.   row = sql.firstRow("select count(*) as numOfRecords from customers where customer_number = ? ", [customerNumber])   
  13.   
  14.   //Verify that customer record exists in Customer Table in the database   
  15.   assert ( 1 == row.numOfRecords ):MSG_CUSTOMER_NOT_FOUND   
  16. }  
  //In a Groovy Assertion Step of a SOAP Request
  def MSG_CUSTOMER_NOT_FOUND = "Customer not found!"
  // Obtain customer number from response of a SOAP Request that creates a customer
  def customerNumber = <... groovy code retrieving customer number from response here ...> 

  //Check if connection to database is available
  if (context.dbConn)
  {
    //connection to the database
    def sql = context.dbConn

    row = sql.firstRow("select count(*) as numOfRecords from customers where customer_number = ? ", [customerNumber])

    //Verify that customer record exists in Customer Table in the database
    assert ( 1 == row.numOfRecords ):MSG_CUSTOMER_NOT_FOUND
  }

4. Finally, in TearDown Script, we close the connection:

  1.   
  2. //In a TearDown Script   
  3. //Close db connection   
  4. if (context.dbConn)   
  5. {   
  6.   context.dbConn.close()   
  7.   log.info "Closed Database Connection."  
  8. }  
//In a TearDown Script
//Close db connection
if (context.dbConn)
{
  context.dbConn.close()
  log.info "Closed Database Connection."
}

Ain’t that simple and reusable? Of course, you can extend this idea to create a connection pool or any other advanced set of objects, map of objects that you can hold onto soapUI context and use whenever needed. However, do not forget to clear up these resources though. That is what the TearDown Script is for!

~srs

2 Responses to “Reusable SQL Connection in soapUI”

  1. Tom Says:

    Hi Sachin,
    I tried this connection below and ran into a problem.

    If I am using an Oracle Thin driver what should the
    dbDriverClassName property value be?

    I am using oracle.jdbc.driver.OracleDriver and also tried jdbc:oracle:thin

    Any ideas?

    Thanks,

    • onebyteatatime Says:

      You can check the Oracle driver documentation in order to figure out exact driver class name. But the class name you just provided seems correct. Have you made sure that you have the driver “jar” in <soapui_home>/bin/ext directory? That is very important. You have to have the driver jar in this directory. Once you start soapUI you shall see this jar file being loaded in console log of soapUI. Once this jar file is loaded at soapUI start up, you should be fine. I think I have documented this point as a comment in my sample script.

      hope that helps.

      ~srs

      Reply

      getting DB connection with Groovy

      Posted by: devakara on: September 8, 2008

      Connecting to DB inside Test Steps for various processes is required most often in any Test Suite. So how do we get a connection of DB2 database (for that matter any database) using Groovy in SOAP UI ?

      Its simple, basically we will get an sql instance first using parameters like JDBC connection URL, driver class name etc..

      This link refers to the Sql class’s API which is used to get the connection

      Say if we use the generic method getInstance(url, user, pwd, driverName) to get the instance and thence execute some SQL queries, we could proceed this way:

      The pre-setup for this would be placing the required jars for DB connectivity in lib folder of SOAP UI and setting their paths to the CLASSPATH variable in bin/soapui.bat file. (In case of DB2 connection, we need to place db2jcc.jar and db2jcc_license_cu.jar jars inside lib directory and set their paths to CLASSPATH variable in soapui.bat file of bin directory). Then,

      1) Import groovy.sql.Sql in the Groovy Step, by including

      import groovy.sql.Sql

      2) Then get the required arguments for Sql.getInstance() method, and fetch the sql instance by

      def  sql = Sql.newInstance(dbPath, dbUserName, dbPassword, dbDriverName);

      Choice of newInstance() method could be as per the options developer has.

      3) Use this sql object for executing the queries to perform operation on DB, for example

      res = sql.execute( “SELECT * FROM TABLE1 WHERE COL1=’123′” );

      26 Responses to "getting DB connection with Groovy"

    • 23 | Igor
      May 19, 2009 at 10:56 pm

      Got it,

      For 2005, the driver changes from
      com.microsoft.sqlserver.jdbc.SQLServerDriver (sql 2000)

      To

      com.microsoft.jdbc.sqlserver.SQLServerDriver

      See: http://blogs.msdn.com/jdbcteam/archive/2007/06/15/java-lang-classnotfoundexception-com-microsoft-jdbc-sqlserver-sqlserverdriver.aspx

      Thanks anyway!
      Igor

package com.hexiang.utils.dao; import java.sql.Connection; import java.sql.DriverManager; import java.sql.ResultSet; import java.sql.SQLException; import java.sql.Statement; import java.util.Properties; import javax.naming.Context; import javax.naming.InitialContext; import javax.naming.NamingException; import javax.sql.DataSource; import org.apache.log4j.Logger; public class DBConnection { /** * 获得与数据库的连接 * * @param path * @return Connection */ public static Connection getConn(String classDriver, String url, String user, String pwd) { try { Class.forName(classDriver); return DriverManager.getConnection(url, user, pwd); } catch (ClassNotFoundException ex) { ex.printStackTrace(); } catch (SQLException ex) { ex.printStackTrace(); } return null; } public static Connection getConn(DataSource dataSource) { try { return dataSource.getConnection(); } catch (SQLException ex) { ex.printStackTrace(); } return null; } public static Connection getConn(String jndiName) { try { Context ctx; ctx = new InitialContext(); DataSource dataSource = (DataSource) ctx.lookup("java:comp/env/" + jndiName); return dataSource.getConnection(); } catch (NamingException ex) { ex.printStackTrace(); } catch (SQLException ex) { ex.printStackTrace(); } return null; } public static Connection getConn(Properties properties) { try { String driver = properties.getProperty("jdbc.driverClassName"); String url = properties.getProperty("jdbc.url"); String user = properties.getProperty("jdbc.username"); String password = properties.getProperty("jdbc.password"); Class.forName(driver); return DriverManager.getConnection(url, user, password); } catch (ClassNotFoundException ex) { ex.printStackTrace(); } catch (SQLException ex) { ex.printStackTrace(); } return null; } /** * oracle连接 * * @param path * @return Connection */ public static Connection getOracleConn(String
""" title: SQL Server Access author: MENG author_urls: - https://github.com/mengvision description: A tool for reading database information and executing SQL queries, supporting multiple databases such as MySQL, PostgreSQL, SQLite, and Oracle. It provides functionalities for listing all tables, describing table schemas, and returning query results in CSV format. A versatile DB Agent for seamless database interactions. required_open_webui_version: 0.5.4 requirements: pymysql, sqlalchemy, cx_Oracle version: 0.1.6 licence: MIT # Changelog ## [0.1.6] - 2025-03-11 ### Added - Added `get_table_indexes` method to retrieve index information for a specific table, supporting MySQL, PostgreSQL, SQLite, and Oracle. - Enhanced metadata capabilities by providing detailed index descriptions (e.g., index name, columns, and type). - Improved documentation to include the new `get_table_indexes` method and its usage examples. - Updated error handling in `get_table_indexes` to provide more detailed feedback for unsupported database types. ## [0.1.5] - 2025-01-20 ### Changed - Updated `list_all_tables` and `table_data_schema` methods to accept `db_name` as a function parameter instead of using `self.valves.db_name`. - Improved flexibility by decoupling database name from class variables, allowing dynamic database selection at runtime. ## [0.1.4] - 2025-01-17 ### Added - Added support for Oracle database using `cx_Oracle` driver. - Added dynamic engine creation in each method to ensure fresh database connections for every operation. - Added support for Oracle-specific queries in `list_all_tables` and `table_data_schema` methods. ### Changed - Moved `self._get_engine()` from `__init__` to individual methods for better flexibility and tool compatibility. - Updated `_get_engine` method to support Oracle database connection URL. - Improved `table_data_schema` method to handle Oracle-specific column metadata. ### Fixed - Fixed potential connection issues by ensuring each method creates its own database engine. - Improved error handling for Oracle-specific queries and edge cases. ## [0.1.3] - 2025-01-17 ### Added - Added support for multiple database types (e.g., MySQL, PostgreSQL, SQLite) using SQLAlchemy. - Added configuration flexibility through environment variables or external configuration files. - Enhanced query security with stricter validation and SQL injection prevention. - Improved error handling with detailed exception messages for better debugging. ### Changed - Replaced `pymysql` with SQLAlchemy for broader database compatibility. - Abstracted database connection logic into a reusable `_get_engine` method. - Updated `table_data_schema` method to support multiple database types. ### Fixed - Fixed potential SQL injection vulnerabilities in query execution. - Improved handling of edge cases in query validation and execution. ## [0.1.2] - 2025-01-16 ### Added - Added support for specifying the database port with a default value of `3306`. - Abstracted database connection logic into a reusable `_get_connection` method. ## [0.1.1] - 2025-01-16 ### Added - Support for additional read-only query types: `SHOW`, `DESCRIBE`, `EXPLAIN`, and `USE`. - Enhanced query validation to block sensitive keywords (e.g., `INSERT`, `UPDATE`, `DELETE`, `CREATE`, `DROP`, `ALTER`). ### Fixed - Improved handling of queries starting with `WITH` (CTE queries). - Fixed case sensitivity issues in query validation. ## [0.1.0] - 2025-01-09 ### Initial Release - Basic functionality for listing tables, describing table schemas, and executing `SELECT` queries. - Query results returned in CSV format. """ import os from typing import List, Dict, Any from pydantic import BaseModel, Field import re from sqlalchemy import create_engine, text from sqlalchemy.engine.base import Engine from sqlalchemy.exc import SQLAlchemyError class Tools: class Valves(BaseModel): db_host: str = Field( default="localhost", description="The host of the database. Replace with your own host.", ) db_user: str = Field( default="admin", description="The username for the database. Replace with your own username.", ) db_password: str = Field( default="admin", description="The password for the database. Replace with your own password.", ) db_name: str = Field( default="db", description="The name of the database. Replace with your own database name.", ) db_port: int = Field( default=3306, # Oracle 默认端口 description="The port of the database. Replace with your own port.", ) db_type: str = Field( default="mysql", description="The type of the database (e.g., mysql, postgresql, sqlite, oracle).", ) def __init__(self): """ Initialize the Tools class with the credentials for the database. """ print("Initializing database tool class") self.citation = True self.valves = Tools.Valves() def _get_engine(self) -> Engine: """ Create and return a database engine using the current configuration. """ if self.valves.db_type == "mysql": db_url = f"mysql+pymysql://{self.valves.db_user}:{self.valves.db_password}@{self.valves.db_host}:{self.valves.db_port}/{self.valves.db_name}" elif self.valves.db_type == "postgresql": db_url = f"postgresql://{self.valves.db_user}:{self.valves.db_password}@{self.valves.db_host}:{self.valves.db_port}/{self.valves.db_name}" elif self.valves.db_type == "sqlite": db_url = f"sqlite:///{self.valves.db_name}" elif self.valves.db_type == "oracle": db_url = f"oracle+cx_oracle://{self.valves.db_user}:{self.valves.db_password}@{self.valves.db_host}:{self.valves.db_port}/?service_name={self.valves.db_name}" else: raise ValueError(f"Unsupported database type: {self.valves.db_type}") return create_engine(db_url) def list_all_tables(self, db_name: str) -> str: """ List all tables in the database. :param db_name: The name of the database. :return: A string containing the names of all tables. """ print("Listing all tables in the database") engine = self._get_engine() # 动态创建引擎 try: with engine.connect() as conn: if self.valves.db_type == "mysql": result = conn.execute(text("SHOW TABLES;")) elif self.valves.db_type == "postgresql": result = conn.execute( text( "SELECT table_name FROM information_schema.tables WHERE table_schema = 'public';" ) ) elif self.valves.db_type == "sqlite": result = conn.execute( text("SELECT name FROM sqlite_master WHERE type='table';") ) elif self.valves.db_type == "oracle": result = conn.execute(text("SELECT table_name FROM user_tables;")) else: return "Unsupported database type." tables = [row[0] for row in result.fetchall()] if tables: return ( "Here is a list of all the tables in the database:\n\n" + "\n".join(tables) ) else: return "No tables found." except SQLAlchemyError as e: return f"Error listing tables: {str(e)}" def get_table_indexes(self, db_name: str, table_name: str) -> str: """ Get the indexes of a specific table in the database. :param db_name: The name of the database. :param table_name: The name of the table. :return: A string describing the indexes of the table. """ print(f"Getting indexes for table: {table_name}") engine = self._get_engine() try: key, cloumn = 0, 1 with engine.connect() as conn: if self.valves.db_type == "mysql": query = text(f"SHOW INDEX FROM {table_name}") key, cloumn = 2, 4 elif self.valves.db_type == "postgresql": query = text( """ SELECT indexname, indexdef FROM pg_indexes WHERE tablename = :table_name; """ ) elif self.valves.db_type == "sqlite": query = text( """ PRAGMA index_list(:table_name); """ ) elif self.valves.db_type == "oracle": query = text( """ SELECT index_name, column_name FROM user_ind_columns WHERE table_name = :table_name; """ ) else: return "Unsupported database type." result = conn.execute(query) indexes = result.fetchall() if not indexes: return f"No indexes found for table: {table_name}" description = f"Indexes for table '{table_name}':\n" for index in indexes: description += f"- {index[key]}: {index[cloumn]}\n" return description # result = conn.execute(query) # description = result.fetchall() # if not description: # return f"No indexes found for table: {table_name}" # column_names = result.keys() # description = f"Query executed successfully. Below is the actual result of the query {query} running against the database in CSV format:\n\n" # description += ",".join(column_names) + "\n" # for row in description: # description += ",".join(map(str, row)) + "\n" # return description except SQLAlchemyError as e: return f"Error getting indexes: {str(e)}" def table_data_schema(self, db_name: str, table_name: str) -> str: """ Describe the schema of a specific table in the database, including column comments. :param db_name: The name of the database. :param table_name: The name of the table to describe. :return: A string describing the data schema of the table. """ print(f"Database: {self.valves.db_name}") print(f"Describing table: {table_name}") engine = self._get_engine() # 动态创建引擎 try: with engine.connect() as conn: if self.valves.db_type == "mysql": query = text( " SELECT COLUMN_NAME, COLUMN_TYPE, IS_NULLABLE, COLUMN_KEY, COLUMN_COMMENT " " FROM INFORMATION_SCHEMA.COLUMNS " f" WHERE TABLE_SCHEMA = '{self.valves.db_name}' AND TABLE_NAME = '{table_name}';" ) elif self.valves.db_type == "postgresql": query = text( """ SELECT column_name, data_type, is_nullable, column_default, '' FROM information_schema.columns WHERE table_name = :table_name; """ ) elif self.valves.db_type == "sqlite": query = text("PRAGMA table_info(:table_name);") elif self.valves.db_type == "oracle": query = text( """ SELECT column_name, data_type, nullable, data_default, comments FROM user_tab_columns LEFT JOIN user_col_comments ON user_tab_columns.table_name = user_col_comments.table_name AND user_tab_columns.column_name = user_col_comments.column_name WHERE user_tab_columns.table_name = :table_name; """ ) else: return "Unsupported database type." # result = conn.execute( # query, {"db_name": db_name, "table_name": table_name} # ) result = conn.execute(query) columns = result.fetchall() if not columns: return f"No such table: {table_name}" description = ( f"Table '{table_name}' in the database has the following columns:\n" ) for column in columns: if self.valves.db_type == "sqlite": column_name, data_type, is_nullable, _, _, _ = column column_comment = "" elif self.valves.db_type == "oracle": ( column_name, data_type, is_nullable, data_default, column_comment, ) = column else: ( column_name, data_type, is_nullable, column_key, column_comment, ) = column description += f"- {column_name} ({data_type})" if is_nullable == "YES" or is_nullable == "Y": description += " [Nullable]" if column_key == "PRI": description += " [Primary Key]" if column_comment: description += f" [Comment: {column_comment}]" description += "\n" return description except SQLAlchemyError as e: return f"Error describing table: {str(e)}" def execute_read_query(self, query: str) -> str: """ Execute a read query and return the result in CSV format. :param query: The SQL query to execute. :return: A string containing the result of the query in CSV format. """ print(f"Executing query: {query}") normalized_query = query.strip().lower() if not re.match( r"^\s*(select|with|show|describe|desc|explain|use)\s", normalized_query ): return "Error: Only read-only queries (SELECT, WITH, SHOW, DESCRIBE, EXPLAIN, USE) are allowed. CREATE, DELETE, INSERT, UPDATE, DROP, and ALTER operations are not permitted." sensitive_keywords = [ "insert", "update", "delete", "create", "drop", "alter", "truncate", "grant", "revoke", "replace", ] for keyword in sensitive_keywords: if re.search(rf"\b{keyword}\b", normalized_query): return f"Error: Query contains a sensitive keyword '{keyword}'. Only read operations are allowed." engine = self._get_engine() # 动态创建引擎 try: with engine.connect() as conn: result = conn.execute(text(query)) rows = result.fetchall() if not rows: return "No data returned from query." column_names = result.keys() csv_data = f"Query executed successfully. Below is the actual result of the query {query} running against the database in CSV format:\n\n" csv_data += ",".join(column_names) + "\n" for row in rows: csv_data += ",".join(map(str, row)) + "\n" return csv_data except SQLAlchemyError as e: return f"Error executing query: {str(e)}" 将上面的工具连接到Microsoft sql server
最新发布
07-23
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值