Oracle® Objects for OLE Developer's Guide 10g Release 2 (10.2) Part Number B14309-01 |
|
|
View PDF |
This quick tour is designed to get you started with Oracle Objects for OLE for Visual Basic. An example application, the employee database application, demonstrates how to program basic database operations, such as navigating through data and, adding, modifying, and querying records. A more advanced section demonstrates how to perform batch inserts using parameter arrays and SQL statement objects. This quick tour and example application assume that the Scott
/Tiger
schema is installed.
The entire code for this example application is provided in the ORACLE_BASE\ORACLE_HOME
\OO4O\VB\SAMPLES\QT\
directory.
This quick tour covers the following topics:
This section introduces the employee database application and the two Visual Basic forms that users interact with to use the application.
The employee database application lets the user do the following:
Browse through data
Add records
Update records
Query the database
Add records in a batch operation
To provide these functions, this example uses the following forms:
The Employee Form displays the fields of the database EMP
table and has functional buttons that allow the user to browse, add, update, and query records.
Figure 6-1 shows the Employee Form.
See Also:
|
The Batch Insert Form allows users to enter records in a batch operation.
See Also: "Programming a Batch Form" for a detailed description of the Batch Insert Form and code for its commands |
Figure 6-2 shows the Batch Insert Form.
Before server data can be manipulated, the application must accomplish the four steps that are described in this section. Sample code for this example is provided in "Completed Sample Form_Load Procedure".
Start the Oracle In-Process Automation Server.
The Oracle In-Process Server (OIP) provides the interface between the Visual Basic application and Oracle Database. To start the Oracle In-Process Server, you must create an OraSession
object using the Visual Basic CreateObject()
function, as follows:
Set OraSession = CreateObject("OracleInProcServer.XOraSession")
When creating the OraSession
object, the argument supplied to the CreateObject()
function must always be OracleInProcServer.XOraSession
. The left side of the argument defines the application name as registered in your system, in this case, OracleInProcServer
. The right side identifies the type of object to create, in this case, the XOraSession
object. Executing this command starts the Oracle In-Process Server.
Connect to Oracle Database.
After the OIP server is running, you can connect to a local or remote Oracle database. To do so, you must create the OraDatabase
object as follows:
Set OraDatabase = OraSession.OpenDatabase("Exampledb", "scott/tiger", _ ORADB_DEFAULT)
The OraSession.OpenDatabase()
method creates the OraDatabase
object. The method call must specify the database name, the connection string, and a bit flag that represents the database mode. The constant ORADB_DEFAULT
represents the default database mode. When Visual Basic executes this line, a connection is created to the specified database.
Create a global OraDynaset
object to manipulate the data.
Oracle Objects for OLE lets users browse and update data using an object called a dynaset.
The Employee application needs a global dynaset that the rest of the program can access. The OraDatabase.CreateDynaset()
method creates the dynaset specifying a valid SQL SELECT
statement. In the example, the statement selects all the rows from the emp
table and assigns the resulting dynaset to the global EmpDynaset
variable as follows:
Set EmpDynaset = OraDatabase.CreateDynaset("select * from emp", _ ORADYN_DEFAULT)
The CreateDynaset()
method returns a pointer to the result of the SQL SELECT statement.
The ORADYN_DEFAULT
parameter value specifies the default dynaset state. In the default state, Oracle Objects for OLE sets unset fields to NULL
while adding records using the AddNew
method. This behavior is preferable because the emp
table has no column defaults defined. You can also specify other options to allow server column defaults when adding records.
Refresh the Employee Form with dynaset data.
The Employee Form displays database records one row at a time. Changes to the current row, such as those caused by navigating to a different row, must be reflected on the screen. The EmpRefresh()
subroutine updates fields with the current dynaset row. For NULL
field values, empty strings are displayed.
The following is an example of an EmpRefresh()
subroutine:
Private Sub EmpRefresh() 'check if the current dynaset row is valid If EmpDynaset.BOF <> True And EmpDynaset.EOF <> True Then txtEmpno = EmpDynaset.Fields("empno").Value ' we can't display nulls, so display "" for NULL fields If Not IsNull(EmpDynaset.Fields("ename").Value) Then txtEname = EmpDynaset.Fields("ename").Value Else txtEname = "" End If If Not IsNull(EmpDynaset.Fields("job").Value) Then txtJob = EmpDynaset.Fields("job").Value Else txtJob = "" End If 'check if mgr=nul If Not IsNull(EmpDynaset.Fields("mgr").Value) Then txtMgr = EmpDynaset.Fields("mgr").Value Else txtMgr = "" End If If Not IsNull(EmpDynaset.Fields("hiredate").Value) Then txtHireDate = EmpDynaset.Fields("hiredate").Value Else txtHireDate = "" End If If Not IsNull(EmpDynaset.Fields("hiredate").Value) Then txtSal = EmpDynaset.Fields("sal").Value Else txtSal = "" End If 'check if comm=nul If Not IsNull(EmpDynaset.Fields("comm").Value) Then txtComm = EmpDynaset.Fields("comm").Value Else txtComm = "" End If txtDeptno = EmpDynaset.Fields("deptno").Value 'if the current dynaset row is invalid, display nothing Else txtEmpno = "" txtEname = "" txtJob = "" txtMgr = "" txtHireDate = "" txtSal = "" txtComm = "" txtDeptno = "" End If End Sub
In the employee application described in the previous section, the Form_Load()
procedure creates the OIP server, connects to the database, creates a global dynaset, and calls the EmpRefresh
function to display the field values on the Employee Form. The following is an example of a Form_Load()
procedure:
Private Sub Form_Load() 'OraSession and OraDatabase are global Set OraSession = CreateObject("OracleInProcServer.XOraSession") Set OraDatabase = OraSession.OpenDatabase("Exampledb", "scott/tiger", 0&) Set EmpDynaset = OraDatabase.CreateDynaset("select * from emp", 0&) Call EmpRefresh End Sub
The following variables must be defined globally in EMP_QT.BAS
:
Global OraSession As Object Global OraDatabase As Object Global EmpDynaset As Object
This section describes the Employee Form in detail and then describes the functions that it uses.
The Employee form displays the fields of the database EMP
table and has functional buttons that allow the user to browse, add, update, and query records.
Each field corresponds to a column in the database EMP
table. The Employee field (ENAME
) is the indexed column and is mandatory for each record. The field data types and sizes are defined as follows in the EMP
table:
Name Null? Type ----------------------- ---------- -------------------------- EMPNO NOT NULL NUMBER(4) ENAME VARCHAR2(10) JOB VARCHAR2(9) MGR NUMBER(4) HIREDATE DATE SAL NUMBER(7,2) COMM NUMBER(7,2) DEPTNO NOT NULL NUMBER(2)
The Employee Number (EMPNO
) and Department (DEPTNO
) columns are NOT
NULL
, and, therefore, always require a value when a record is added. The length of each field is enforced by setting the MaxLength
property of each TextBox
to the appropriate number.
Figure 6-3 shows the Employee Form.
The initial code for the actual Form_Load
procedure is provided in "Completed Sample Form_Load Procedure".
The Employee form is initialized by the Form_Load()
procedure and includes the following features:
Database applications typically require that the user be able to view data in the database. The Employee form has four buttons that let the user scroll through data. Table 6-1 lists the buttons, what they do, which dynaset move method enables the action of the button, and where to look for further information.
Table 6-1 Navigational Buttons and Dynaset Move Methods
Button | Action | Method | See... |
---|---|---|---|
|< |
Moves to the first record | MoveFirst |
Moving to First or Last Rows |
< |
Moves to the previous record | MovePrevious |
Moving to the Previous Row |
> |
Moves to the next record | MoveNext |
Moving to the Next Row |
>| |
Moves to the last record | MoveLast |
Moving to First or Last Rows |
To enable navigation through the records of the Employee database, you must first create a global dynaset that selects all the records (rows). Then use the dynaset move methods to program the navigation buttons.
To enable a move to the first row of a dynaset, use the MoveFirst
method. Then call the EmpRefresh()
routine to refresh the data in the Employee form.
The following example code shows the first-click event procedure for the employee example:
Private Sub cmdFirst_Click() EmpDynaset.MoveFirst Call EmpRefresh End Sub
For a move to the last row, use the MoveLast
method. Then, call the EmpRefresh()
routine to refresh the data in the Employee form.
The following example code shows the last-click event procedure for the employee example:
Private Sub cmdLast_Click() EmpDynaset.MoveLast Call EmpRefresh End Sub
Navigation is possible to any row of a dynaset. If a user is positioned in the middle of a dynaset (that is, the current row is not the first row), the MovePrevious
method enables navigation to the previous row.
However, when a user is positioned on the first row (current row is the first row) and executes the MovePrevious
method, the beginning-of-file (BOF) condition becomes TRUE
and the current row becomes invalid. In this case, the current row must be reset to the first row using the MoveFirst
method.
The following example code shows the click-event procedure for the Previous button:
Private Sub cmdPrevious_Click() If EmpDynaset.BOF <> True Then EmpDynaset.DbMovePrevious If EmpDynaset.BOF = True Then MsgBox WarnFirstEmp$ EmpDynaset.DbMoveFirst End If End If
If a user is positioned in the middle of a dynaset (that is, the current row is not the last row), the MoveNext
method enables navigation to the next row.
However, when a user is positioned on the last row (current row is the last row) and then executes MoveNext
, the end-of-file condition (EOF) becomes TRUE
and the current row becomes invalid. In this case, the current row must be reset to the last row using the MoveLast
method.
The following example code shows the click-event procedure for the Next button:
Private Sub cmdNext_Click() If EmpDynaset.EOF <> True Then EmpDynaset.DbMoveNext If EmpDynaset.EOF = True Then MsgBox WarnLastEmp$ EmpDynaset.DbMoveLast End If End If
In the example application, the following buttons allow users to add employee records to the database:
Add
Commit
To add a record, the user clicks on the Add button, enters the new fields in the text boxes, and then clicks the Commit button to save the data to the database.
The Add event procedure must perform the following steps:
Clear the fields on the form.
Disable the Add button.
Enable the Commit button.
Let the user enter new field values.
The following example code shows the Add event procedure for the Add button:
Private Sub AddNew_Click() 'Blank out the fields txtEmpno = "" txtEname = "" txtJob = "" txtMgr = "" txtHireDate = "" txtSal = "" txtComm = "" txtDeptno = "" 'Disable the Add button and enable the commit button AddNew.Enabled = False Commit.Enabled = True 'Disable the navigation buttons DisableNavButtons 'Set doadd to true for commit procedure DoAdd = True End Sub
When the AddNew_Click()
method exits, control returns to the Employee Form where the user enters values in the fields.
To commit an addition, you must place the dynaset in add mode using the AddNew
method. Then, you assign the new data to the dynaset fields and update the database using the Update
method. To make the program robust, the software validates some fields before adding them to the database.
The Commit_Click()
event procedure for adding records must do the following:
Check that the Employee Number and Department fields are not null.
Check that the new Employee Number is not a duplicate entry.
Steps 1 and 2 are performed by the DoValidationChecks()
function which is described following the Commit_Click()
.
Place the dynaset in add mode using the AddNew
method.
Assign entered data to dynaset fields using the Fields().Value
property. This step is performed by the UpdateDynasetFields
function.
Update the database with new records, using the Update
method.
Disable the Commit button.
Enable the Add button.
The code for the Commit
function is broken into the following routines:
The following is a typical Commit_Click()
event procedure for adding records:
Private Sub Commit_Click() On Error GoTo err_commit ErrMsg = "" 'Do validation checks on entered data If DoValidationChecks Then 'If validation checks have passed 'Add the new record to dynaset EmpDynaset.AddNew 'Update the dynaset fields and then update database if there is no error. If UpdateDynasetFields Then 'Update the database EmpDynaset.Update Commit.Enabled = False AddNew.Enabled = True Exit Sub err_commit: If ErrMsg <> "" Then MsgBox ErrMsg Else MsgBox Error$ End If End Sub
To check for duplicate entries as suggested in Step 2, you must create a local dynaset with the NOCACHE
option, using a SQL statement that counts the rows matching the entered Employee Number field. If a match is found (row count greater than 0), the entered employee number is a duplicate entry and an error is displayed. In this case, because the SQL SELECT
statement returns only a number, creating the dynaset without a cache is a more efficient error check than the server finding a duplicate entery.
DoValidationChecks()
returns True
if the entered data is valid; otherwise, it returns False
.
Function DoValidationChecks() As Boolean Dim DupDyn As Object Dim DupDynQry As String On Error GoTo err_ValidationCheck ErrMsg = "" 'Empno cannot be changed while in Update mode, so we can skip over validation If DoAdd Then If txtEmpno = "" Then ErrMsg = "You must enter a value for Employee Number" Error 1 End If End If If txtHireDate <> "" And Not IsDate(txtHireDate) Then ErrMsg = "Enter date as dd-mmm-yy." Error 2 End If If txtDeptno = "" Then ErrMsg = "You must enter a value for Department Number" Error 3 End If 'If adding a record, check for Duplicate empno value by 'attempting to count rows with same value 'Build Query: If DoAdd Then DupDynQry = "select count(*) from emp where empno = " & txtEmpno Set DupDyn = OraDatabase.CreateDynaset(DupDynQry, ORADYN_NOCACHE) If DupDyn.Fields(0).Value <> 0 Then ErrNum = DUPLICATE_KEY ErrMsg = "Employee Number already exists." Error ErrNum End If End If 'Succesful validation with no errors returns True DoValidationChecks = True Exit Function err_ValidationCheck: If ErrMsg <> "" Then MsgBox ErrMsg Else MsgBox Error$ End If 'Validation returns false on failure DoValidationChecks = False End Function
The commit event procedure calls this function after putting the dynaset in either Edit
or AddNew
mode. The UpdateDynasetFields()
function sets the dynaset fields to the values entered in the text boxes. The function returns TRUE
if successful, or returns FALSE
if there is an error.
Function UpdateDynasetFields() As Integer 'This function sets the dynaset field value to those entered in the text boxes. 'The function returns true on success, false on error. ErrMsg = "" On Error GoTo err_updatedynasetfields EmpDynaset.Fields("empno").Value = txtEmpno EmpDynaset.Fields("ename").Value = txtEname EmpDynaset.Fields("job").Value = txtJob EmpDynaset.Fields("mgr").Value = txtManager EmpDynaset.Fields("hiredate").Value = txtHireDate EmpDynaset.Fields("sal").Value = txtSal EmpDynaset.Fields("comm").Value = txtComm EmpDynaset.Fields("deptno").Value = txtDeptno UpdateDynasetFields = True Exit Function err_updatedynasetfields: If ErrMsg <> "" Then MsgBox ErrMsg Else MsgBox Error$ End If UpdateDynasetFields = False
To allow users to update existing records in the database, you need to include an Update button in the Employee Form. Users navigate to a particular record, click the Update button, make changes, and then click the Commit button.
While in update mode, the application makes the following restrictions:
Users cannot navigate to another record or perform another function.
Users cannot change the employee number because this is the primary key.
To program the Update function, write an event procedure for the Update button and modify the Commit procedure so that it handles both updating and adding records.
To code the Update button, disable the Employee Number text box to prevent changes to this field while updating records, because this is a primary key. You must also disable the other buttons to disable other functions, such as navigation, while updating records.
Set the DoUpdate
Boolean expression to TRUE
, so the commit procedure recognizes the current process as an update operation, not an addition.
The update event procedure must do the following:
Disable the Update button.
Enable the Commit button.
Disable other buttons to disable functions, such as navigation, during the update operation.
Disable the Employee Number text box.
Set the DoUpdate
flag to True
.
Let the user enter changes.
The following example code shows the update event procedure:
Private Sub cmdUpdate_Click() 'Disable the Update button and enable the commit button cmdUpdate.Enabled = False Commit.Enabled = True 'Disable all other buttons DisableNavButtons txtEmpno.Enabled = False DoUpdate = True End Sub
The update and add event procedures call the DisableNavButtons()
subroutine to disable navigation and other functions during an add or update operation.
Private Sub DisableNavButtons() 'disable all buttons while adding and updating cmdFirst.Enabled = False cmdPrevious.Enabled = False cmdNext.Enabled = False cmdLast.Enabled = False cmdFind.Enabled = False cmdUpdate.Enabled = False AddNew.Enabled = False End Sub
The procedure for committing an update operation is similar to committing an add, except that the dynaset is set in edit mode using the Edit
method and then the new dynaset values are assigned.
Because the same commit button and the same commit event procedure are used to add and update, two global flags DoAdd
and DoUpdate
are added to distinguish between adding and updating. The Add and Update click event procedures set these flags.
The Commit event procedure for adding and updating must do the following:
Validate entered data using the DoValidationChecks()
function as before.
Use AddNew
to add records or else use Edit
for updates.
Assign entered data to dynaset fields, using the Fields().Value
property using UpdateDynasetFields()
as before.
Update database with new records, using Update
.
Disable the Commit button.
Reenable all other functional buttons including the Add and Update buttons.
Set the DoUpdate
and DoAdd
flags to False
.
The code that changes button and flag states in Steps 5 through 7 is provided in a new subroutine called SetAfterCommitFlags()
. This replaces the lines of code that originally enabled Commit
and AddNew
.
The code for this Commit function is broken into the following routines:
"DoValidationChecks( ) Function" (UNKNOWN STEP NUMBER) , also used in the original Commit
function
"UpdateDynasetFields( ) Function", also used in the original Commit
function
"SetAfterCommitFlags() Subroutine Example", which is a new subroutine
The following example shows the Commit_Click
Event Procedure.
Private Sub Commit_Click() On Error GoTo err_commit ErrMsg = "" 'Do validation checks on entered data If DoValidationChecks Then 'If validation checks have passed 'If we are adding a record use AddNew If DoAdd = True Then EmpDynaset.AddNew End If 'If we are updating a record use Edit If DoUpdate = True Then EmpDynaset.Edit End If 'Update the dynaset fields and then update database if there is no error. If UpdateDynasetFields Then EmpDynaset.Update End If SetAfterCommitFlags End If 'Endif for DoValidationChecks Exit Sub err_commit: If ErrMsg <> "" Then MsgBox ErrMsg Else MsgBox Error$ End If End Sub
The following example shows the SetAfterCommitFlag()
Subroutine.
The SetAfterCommitFlags()
subroutine is called at the end of the commit event procedure. The SetAfterCommitFlags()
subroutine reenables disabled buttons and text boxes and sets the DoUpdate
and DoAdd
flags to False
.
Sub SetAfterCommitFlags() 'disable commit and re-enable add and update buttons Commit.Enabled = False AddNew.Enabled = True cmdUpdate.Enabled = True 'enable the other buttons cmdFirst.Enabled = True cmdPrevious.Enabled = True cmdNext.Enabled = True cmdLast.Enabled = True cmdFind.Enabled = True cmdUpdate.Enabled = True AddNew.Enabled = True DoUpdate = False DoAdd = False txtEmpno.Enabled = True End Sub
Users can delete records by navigating to a particular record and clicking the Delete button. The application prompts the user to verify the deletion, then the application deletes the record using the Delete
method. The program then refreshes the screen with the next record or with the previous record if the user deleted the last record in the dynaset.
The following example shows the delete-click event procedure:
Private Sub cmdDelete_Click() 'prompt user Response = MsgBox("Do you really want to Delete?", vbYesNo + vbExclamation) If Response = vbYes Then EmpDynaset.Delete 'attempt to move to next record EmpDynaset.MoveNext If EmpDynaset.EOF Then 'If deleted last record EmpDynaset.MovePrevious End If Call EmpRefresh End If End Sub
The employee application can be configured to allow users to search for particular records in the database. For demonstration purposes, a Find button is included to allow users to query only employee names. At any time, the user can enter the query in the Employee Name field, and click the Find button. The application then displays the result or displays a message if the name cannot be found.
To search for records, the FindFirst
method is used. When the find operation succeeds, the record is displayed. If the find fails, a message is displayed. The current row is reset to the first row, because failures cause the dynaset to be BOF (beginning-of-file), effectively making the current row invalid.
The Find_Click()
event procedure must do the following:
Build a find clause to find the record where the ENAME
column matches the entered string.
Execute the find using the FindFirst
method.
Display the record if it is found; if the record was not found, display a message and reset the current row to the first row.
The following example shows a typical find click event procedure:
Private Sub cmdFind_Click() Dim FindClause As String Dim SingleQuote As String ErrMsg = "" SingleQuote = "'" On Error GoTo err_find 'build the find clause: 'Can make our query case insensitive by converting the names to upper case 'FindClause = "UPPER(ename) = " & SingleQuote & UCase(txtEname) & SingleQuote FindClause = "ename = " & SingleQuote & txtEname & SingleQuote EmpDynaset.DbFindFirst FindClause If EmpDynaset.NoMatch Then MsgBox "Could not find record matching Employee Name " & txtEname EmpDynaset.DbMoveFirst End If Call EmpRefresh Exit Sub
This section describes the Batch Insert Form and then describes the functions that it uses.
The Batch Insert Form allows users to insert rows in a batch operation, that is, to insert more than one record into the database by using only one command. This feature is implemented using parameter arrays and SQL statements.
Table 6-1 shows a typical Batch Insert Form:
Users navigate to the Batch Insert Form by clicking the Batch Insert button on the Employee Form. The Batch Insert Form has a grid that displays the entered data and a row of fields where the user enters each record. To keep the example simple, users are only allowed to enter information into the Employee Number, Employee Name, and Department Number fields.
Users enter records in the fields and click the Add to Grid button. The program displays the entered records in the grid. To insert the entire batch to the database, users click the CommitGrid button.
The Batch Insert Form uses three procedures. The Form_Load()
procedure initializes the grid with the column headers. The CmdAddtoGrid_click()
procedure copies the entered data from the fields to the grid. The CommitGrid_Click()
procedure contains the parameter array and SQL statements used to make the batch insert.
These procedures are described as follows:
The following examples show how the Batch Insert Form_Load()
procedure sets the column headings for the grid:
Private Sub Form_Load() Grid1.Enabled = True CurrRow = 0 'Top row ReadRow = 0 ReadCol = 0 'Set column headings Grid1.Row = CurrRow Grid1.Col = 0 Grid1.Text = "Employee Number" Grid1.Col = 1 Grid1.Text = "Employee Name" Grid1.Col = 2 Grid1.Text = "Department Number" NoOfCols = 3 CurrRow = CurrRow + 1 End Sub
The CmdAddtoGrid_Click()
procedure copies the data entered in the fields to the next empty grid row. The global variable CurrRow
always points to the first empty row in the grid.
The following example shows the CmdAddtoGrid_Click()
:
Private Sub CmdAddtoGrid_Click() 'Update the grid 'Update Empno column Grid1.Row = CurrRow Grid1.Col = 0 Grid1.Text = txtEmpno 'Update Ename column Grid1.Row = CurrRow Grid1.Col = 1 Grid1.Text = txtEname 'Update Deptno column Grid1.Row = CurrRow Grid1.Col = 2 Grid1.Text = txtDeptno 'Increment CurrCol CurrRow = CurrRow + 1 NoOfRows = CurrRow - 1 End Sub
The CommitGrid_Click()
procedure inserts the grid data into the database. To do so, this procedure creates a parameter array object for each column in the EMP
table that corresponds to a column in the grid. The OraParameters.AddTable()
method defines each parameter array. For example, a parameter array called EMPNO_ARR
holds all Employee Number column elements.
After the parameter arrays are defined, the Put_Value
method populates them with grid column elements.
To commit the parameter array elements to the database, this procedure uses the CreateSQL()
method with a SQL INSERT
statement containing the parameter arrays. Because the CreateSQL()
method executes the SQL INSERT
statement in addition to creating a SQL statement object, all column elements (parameter array elements) are inserted into the EMP
table with this one statement.
If an error occurs during a SQL INSERT
statement that contains parameter arrays, the SQL statement object is still created with no explicitly raised error. To identify such errors, always check the OraDatabase.LastServerErr
and OraDatabase.LastServerErrText
properties immediately after executing the CreateSQL
method.
The CreateSQL
method updates the database directly and has no effect on the dynaset. The EmpDynaset
. Refresh
method must used to refresh this dynaset so that it reflects the newly inserted records.
The CommitGrid_Click()
event procedure must do the following:
Define a parameter array for each grid (database) column, using the AddTable
method.
Copy grid column elements into parameter arrays, using the Put_Value
method within a nested loop.
Create a SQL statement object using the CreateSQL
method to insert parameter array elements into the EMP
table.
Check the LastServerErrText
and LastServerErr
properties to catch SQL statement execution errors.
Refresh the global dyanset to reflect newly inserted records, using the Refresh
method.
The following example shows a typical cmdCommitGrid_Click()
procedure:
Private Sub cmdCommitGrid_Click() Dim OraSqlStmt As Object Dim OraPArray(2) As Object On Error GoTo err_CommitGrid ErrMsg = "" 'Define parameter arrays, one for each column OraDatabase.Parameters.AddTable "EMPNO_ARR", ORAPARM_INPUT, ORATYPE_NUMBER, _ NoOfRows OraDatabase.Parameters.AddTable "ENAME_ARR", ORAPARM_INPUT, ORATYPE_VARCHAR2, _ NoOfRows, 10 OraDatabase.Parameters.AddTable "DEPTNO_ARR", ORAPARM_INPUT, ORATYPE_NUMBER, _ NoOfRows If OraDatabase.LastServerErr <> 0 Or OraDatabase.LastServerErrText <> "" Then Error 1 End If 'Initialize local array to hold parameter arrays Set OraPArray(0) = OraDatabase.Parameters("EMPNO_ARR") Set OraPArray(1) = OraDatabase.Parameters("ENAME_ARR") Set OraPArray(2) = OraDatabase.Parameters("DEPTNO_ARR") 'Init the param array variables. Add loop to read thru grid ROWS For ReadRow = 0 To (NoOfRows - 1) Grid1.Row = ReadRow + 1 'Loop to read thru grid CELLS For ReadCol = 0 To NoOfCols - 1 Grid1.Col = ReadCol OraPArray(ReadCol).Put_Value Grid1.Text, ReadRow Next ReadCol Next ReadRow 'create a sqlstmt to insert array values into table Set OraSqlStmt = OraDatabase.CreateSql("insert into emp(empno,ename,deptno)" & _ "values(:EMPNO_ARR,:ENAME_ARR,:DEPTNO_ARR)", 0&) If OraDatabase.LastServerErr <> 0 Or OraDatabase.LastServerErrText <> "" Then ErrMsg = OraDatabase.LastServerErrText Error 1 End If 'Refresh the Dynaset EmpDynaset.Refresh OraDatabase.Parameters.Remove "EMPNO_ARR" OraDatabase.Parameters.Remove "ENAME_ARR" OraDatabase.Parameters.Remove "DEPTNO_ARR" Exit Sub err_CommitGrid: If ErrMsg <> "" Then MsgBox ErrMsg Else MsgBox Error$ End If End Sub