Collect
Set up Connection Types
Connection Types require SQL Server to download the data dictionary from the source database, a substring sample command and the windows driver to embed in the package. Columns included in the download must be named the same as the example in the database. The connection string in this table must be fully populated or consequently, the database user ID and password are ignored when using a user-defined ODBC connection.
To add a connection type:
- Select Administrative > Connection Types in Navigation pane.
-
Click Sources for SQL SERVER Target Connection Type.
NOTE: Collect only supports SQL Server Source databases.
-
Click Add.
View the field descriptions for the Source Connection Types page
-
Enter the connection name in SOURCE CONNECTION TYPE field.
NOTE: The SOURCE CONNECTION TYPE is the specific DBMS to which Collect connects. In order for the correct database transaction to called, the SOURCE CONNECTION TYPE must begin with “ODBC,” “SQLSERVER,” “ORACLE” or “OLE” followed by any combination of letters and numbers.
- Enter a provider in OLE PROVIDER field or enter a space if using an ODBC driver.
- Enter a driver in ODBC DRIVER field or enter a space if using an OLE provider.
- Click Save.
- Click Vertical View to further configure the Source Connection Type.
-
Click Edit.
View the field descriptions for the Source Connection Types page's Vertical View
- Use the following table to populate the applicable connection settings based on DBMS requirements:
Field |
Description |
Fetchsize |
Enter value for the number of records put into a buffer on the Windows server. |
Download SQL |
Enter SQL command used to read non-SAP databases to download table and column information from the source during the Build Package process. The data type conversions are performed during this load process and followed by Create Table and Create Keys/Indices. |
Download Table SQL |
Enter SQL called during the Fetch Table process. A list of tables that is available to download is pulled from the Source system. |
Download Primary Key SQL |
Enter SQL called during the Build Package process. SQL verifies the table built from SQL has a primary key. If this field is blank, the new table in SQL Server will not have any primary keys. |
Download Indices SQL |
Enter SQL called during the Build Package process. SQL verifies the table built from SQL has indices. This field is only used for non-SAP sources. If this field is blank, the new table in SQL Server will not have indices. |
SAP Table Check SQL |
Enter SQL called during the Build Package process to verify if the table is Transparent, Pooled, Clustered or Other. Tables are not created for objects (views, structures, etc.) marked as “other.” Pooled or Clustered tables must use a package type of BOA RFC or SAP RFC. Transparent tables must use a package type of CranPort, SSIS or Data Services. |
SAP Column Download SQL |
Enter SQL called during the Build Package process. Downloads all column information from the SAP DD03L table. |
SAP Primary Key SQL |
Enter SQL called during the Build Package process to download all column information from SAP DD03L table. |
ODBC Driver |
Enter name of driver used in the Windows ODBC setup for the DSN. The driver is used inside packages when performing data downloads. Collect uses .Net connections for performing tasks, but the packages still use these drivers. The ODBC driver for Oracle and SQL Server can remain blank because the default drivers are pre-built in the systems for the applicable DBMS. |
OLE Provider |
Enter OLE settings. Only used for OLE type of connections. Oracle and DB2 have better support for Unicode characters using OLE connection type. |
Connection String Example |
Enter an example of information for clients to reference. |
Left Column Delimiter |
Enter character that wraps around special characters in column and table names. SAP uses slashes in table names, which need to be wrapped in delimiters to prevent SQL errors. |
Right Column Delimiter |
Enter character that wraps around special characters in column and table names. SAP uses slashes in table names, which need to be wrapped in delimiters to prevent SQL errors. |
- Click Save.
- Close Vertical View.
Data Type Conversions
When you create connection types in Collect, you can set and update data type conversions using the Connection Type Data Conversion page. You can apply validation rules that are used to ignore, error out, or accept specific data type conversions from a Source to a Target.
Watch Video: Set Up and Update Data Type Conversions
To Set Up a Data Type Conversion:
- Click Data Type Conversions for the Source Connection Type to configure the data type conversions for the connection type.
Click Add.
View the field descriptions for the Connection Type Data Conversion page
Enter a sort value in PROCESS ORDER field to determine the order in which the Data Type validation rules are processed.
NOTE: Syniti recommends having all error conditions first followed by accept conditions. The default can be set to accept any other data type or to error out any data type not found in the data type condition process.
- Select a validation rule from DATA VALIDATE CONTROL list box to validate data. Values are:
-
Any Scale Greater Than 0
- Data Type and Length Match
- Data Type and Scale Match
- Data Type Match Only
- Data Type, Length and Scale Match
- Data Type, Precision and Scale Match
- Default for any Data Type
- Greater Than or Equal Maximum Length
- Select a flag from PROCESS FLAG list box to determine how data is processed. Values are:
-
Accept – accepts data validate controls
-
Error – errors out process
-
Ignore – columns of defined data type are ignored when building a package.
- Enter a type (from source) in DATA TYPE field, e.g., Image, Text, Strg, or Any.
- Enter a length (from source) in DATA LENGTH field. Enter 0 if the field does not require a length.
-
Enter a type format (from target) in NEW DATA TYPE FORMAT field.
NOTE: The NEW DATA TYPE FORMAT is the format for the new type of data to be defined in the target. It is defined in SQL Server and must contain the length or the scale.
-
Click Save; the Vertical View displays.
View the field descriptions for the Connection Type Data Conversion page's Vertical View
-
Enter a precision, the total number of digits (from source), in Data Precision field. Enter 0 if the field is not a numeric field.
NOTE: Data Precision only applies to numbers, but the total digit count of the number and this value match to the download data.
-
Enter a scale, the number of digits after decimal point (from source), in Data Scale field. Enter 0 if field is not a decimal field.
NOTE: Data Scale applies to numbers containing decimals. This field must match the downloaded number of decimals.
- Enter a type in New Data Type field for the field in target, e.g., Image, Text, Strg, or Any.
- Enter a length in New Data Length field for the field in target. Enter 0 if the field does not require a length.
-
Enter a precision (total number of digits) in New Data Precision field for the field in target. Enter 0 if the field is not a numeric field.
NOTE: New Data Precision only applies to numbers, but the total digit count of the number and this value matches to the download data.
-
Enter a scale (number of digits after decimal point) in New Data Scale field for the field in target. Enter 0 if field is not a decimal field.
NOTE:New Data Scale is the scale of the decimal being created.
-
Enter a template in Column Override Template field.
NOTE: The Column Override Template is used to convert a source column into a new format in the target database. Enter 0 if the field is not a decimal.
- Click Save.
SQL commands used to read the DBMS tables for table and column information are stored in Collect. The Data Types have been customized for each DBMS in combination with SAP data types defined in DD03L. In order to support loading data in the Target database, the packages must be manually built for connections that do not have system tables. For example, Access and Top Speed are file-based storage databases. Excel spreadsheet data can be imported via a manual package.
SQL Server and Oracle are connection types built into Collect. There is a process to build customized ODBC connections against Sybase, Informix, DB2, etc. A customized ODBC connection uses the connection string only in the Connection table and must be completely populated with the user ID and password.