Insert JDBC Data in Bulk Assertion
The Insert JDBC Data in Bulk assertion is designed to add a large amount of data into a JDBC data source. It parses data from (by default) a comma-separated value (CSV) file and then inserts them in bulk into a specified database using a JDBC connection. The CSV file is obtained from the source message. For information on how to set the message source, see .
gateway91
The
Insert JDBC Data in Bulk
assertion is designed to add a large amount of data into a JDBC data source. It parses data from (by default) a comma-separated value (CSV) file and then inserts them in bulk into a specified database using a JDBC connection. The CSV file is obtained from the source message. For information on how to set the message source, see Select a Target Message. Throughout this topic, "CSV" is used to denote a comma-separated value file. However, comma separators are only the default. Any character may be used as the field delimiter.
For inserting large amounts of data, this assertion offers better performance than using the Perform JDBC Query Assertion, which is best suited for inserting smaller amounts of data.
Context Variables
None.
Cluster Properties
None.
Properties
Setting | What you should know... |
JDBC Connection | Select a preconfigured connection from the drop-down list or specify a context variable. |
Table Name | Enter the name of the database table to update. You may specify context variables. |
Record Delimiter | Select the file record delimiter used in the CSV file. The default is CRLF . |
Field Delimiter | Optionally specify the field record delimiter used in the CSV file. Must be a single character. You may specify context variables. The default is , (comma). |
Quoted? | Select the check box if the CSV file contains quotes. Otherwise, the document is parsed without enforcing the quotes. |
Quote Char | If quotes are used, optionally specify the quote character. You may specify context variables. The default is " (double quotes). |
Escape Char | If quotes are used, optionally specify the escape character. You may specify context variables.The default is empty. |
Data mapping table | This table describes how to map the data from the CSV file to the database table. Note that the CSV field number starts from 0. |
Decompression | Specify the compression type used to compress the data. |
Batch Size | Specify the batch size for data to be inserted into the database. The default is 100 . |
Table Mapping Properties
Setting | What you should know... |
Table Column | Specify the database table column where the values will be inserted. |
CSV Field | Specify the field number (starting from '0') in the CSV file from which to get the value. |
Transformation | Select the transformation to be performed on the value. The transformations currently supported:
|
Parameter | If the selected transformation accepts a parameter, enter it here. |
Frequently Asked Questions
Question | Answer |
How can I tell if data was inserted correctly into the database? | Look for these audit entries: -4, -3. |
How can I tell if this assertion failed? | Look for these audit entries: 2, -5. |
Where is the CSV file located? | The CSV data is obtained from the target message (request, response, or context variable). It can also be added via a REST API call. |