Skip to content

elasticio/jdbc-component

Repository files navigation

CircleCI

JDBC-component

Table of Contents

Execute stored procedure

General information

Description

This is an open source component for working with object-relational database management systems on elastic.io platform.

Completeness Matrix

JDBC Component Completeness Matrix

JDBC Component Completeness Matrix

Credentials

You need to use following properties to configure credentials:

DB Engine - Choose one of existing database types: image

Connection URI - Provide hostname of the server, e.g. acme.com

Connection port - Optional field. Provide port of the server instance, as by default:

  • 3306 - MySQL
  • 5432 - PostgreSQL
  • 1521 - Oracle
  • 1433 - MSSQL

Database Name - Provide name of database at the instance that you want to interact with.

User - Provide a username that has permissions to interact with the Database.

Password - Provide a password of the user that has permissions to interact with the Database.

Configuration properties - Optional field. Provide a configuration properties for connections to the Database, e.g. useUnicode=true&serverTimezone=UTC

Limitation: Configuration properties value may not be checked during Credentials Verifacation, so in case of using this field make sure that it contains correct input.

Triggers

Select trigger

You are able to provide SELECT query with last execution timestamp as WHERE clause criteria. image Before executing the the statement %%EIO_LAST_POLL%% will be replaced with ISO Date of the last execution or max value of the last pooled datetime, for example 2018-08-01T00:00:00.000. During the first execution, date will be equal to "start" of Unix Time - 1970-01-01 00:00:00.000. Precision of the polling clause can be till milliseconds. The format of Start Polling From (optional) field should be like yyyy-mm-dd hh:mi:ss[.sss], where

  • yyyy - year
  • mm - month
  • dd - day
  • hh - hour
  • mi - minute
  • ss - second
  • sss - millisecond (optional)

Get Rows Polling trigger

This trigger can polling data from provided table. As WHERE clause you can use column, which has datatype like DATE or TIMESTAMP. image Before executing the the statement %%EIO_LAST_POLL%% will be replaced with ISO Date of the last execution or max value of the last pooled datetime, for example 2018-08-01T00:00:00.000. During the first execution, date will be equal to "start" of Unix Time - 1970-01-01 00:00:00.000. Precision of the polling clause can be till milliseconds. The format of Start Polling From (optional) field should be like yyyy-mm-dd hh:mi:ss[.sss], where

  • yyyy - year
  • mm - month
  • dd - day
  • hh - hour
  • mi - minute
  • ss - second
  • sss - millisecond (optional)

Please Note: Component Snapshot will not be overwritten in Real-Time flows due to platform behaviour, so we strongly recommend to use Get Rows Polling trigger in Keen Flows only

Input fields description

image

Tables List

Dropdown list with available table names, required field

Timestamp (or similar) Column

Dropdown list with available Column names, that have a type like java.sql.Date or java.sql.Timestamp, required field

Start Polling From (optional)

Optional field, indicates the beginning time to start polling from (defaults to the current time)

SELECT trigger (Deprecated)

This action exists in JDBC component only for backward compatibility. New Select trigger is recommended to use.

Actions

Execute custom query

Action to execute custom SQL query from provided request string.

Note: SQL request will be executed according to chosen database JDBC specification.

Execution result returns as array of objects. If request contains multiple sql statements - them will execute inside one transaction. If one of statements fails, transaction will be rollbacked.

Input fields description

As input metadata, you will get one field named query to provide request string

Query Samples:

Select:

SELECT name, size FROM stars

Update:

INSERT INTO stars values (1,'Taurus', '2015-02-19 10:10:10.0', 123, 5, 'true', '2015-02-19')

Posgresql batch multiple statements request:

DELETE FROM stars WHERE id = 1;
UPDATE stars SET radius = 5 WHERE id = 2;

Select action

image

The action will execute an SQL query that can return multiple results, it has limitations on the query and suited only for SELECT type of queries. In SQL query you can use clause variables with specific data types. Internally we use prepared statements, so all incoming data is validated against SQL injection, however we had to build a connection from JavaScript types to the SQL data types therefore when doing a prepared statements, you would need to add :type to each prepared statement variable.

Note: prepared statement variables name could contain: any characters between a-z or A-Z, a digit and a character _ ([a-zA-Z0-9_]).

For example if you have a following SQL statement:

SELECT
FROM users
WHERE userid = @id AND language = @lang

you should add :type to each @parameter so your SQL query will looks like this:

SELECT
FROM users
WHERE userid = @id:number AND language = @lang:string

Following types are supported:

  • string
  • number
  • bigint
  • boolean
  • float
  • date

image

Dropdown Emit Behaviour contains following possible options:

  • Fetch all - a single message with an array results containing all the objects (rows) will be emitted
  • Emit Individually - multiple messages (one message per one row) will be emitted
  • Expect Single - a single message with one result row will be emitted. If more than one row is returned the error will be thrown. A boolean input "Allow Zero Results" (defaults to false) appears at input metadata. If false - error will be thrown, else - the empty object will be emitted.

image

Input fields description

Component supports dynamic incoming metadata - as soon as your query is in place it will be parsed and incoming metadata will be generated accordingly.

Lookup Row By Primary Key

image

The action will execute select query from a Table dropdown field, as criteria can be used only PRIMARY KEY. The action returns only one result (a primary key is unique). Checkbox Don't throw Error on an Empty Result allows to emit an empty response, otherwise you will get an error on empty response.

Input fields description

image As an input metadata you will get a Primary Key field to provide the data inside as a clause value.

Insert action

The action will execute INSERT command into the table from Table dropdown list the values specified in the body.

List of Expected Config fields

  • Enable Rebound if Yes in case of deadlocks rebound message using Sailor rebound mechanism, number of rebound can be specified via environment variable: ELASTICIO_REBOUND_LIMIT recommended value 3

Input fields description

Table

Action contains only one configuration field Table - dropdown list with available table names. image

Expected input metadata

As input metadata, you will get all fields of the selected table except for fields with auto-increment or auto-calculated property.

Expected output metadata

As output metadata, you will get execution insert result like:

{
  "result": true
}

Delete Row By Primary Key

image The action will execute delete query from a Table dropdown field, as criteria can be used only PRIMARY KEY. The action returns count of affected rows. Checkbox Don't throw Error on an Empty Result allows to emit an empty response, otherwise you will get an error on empty response. Enable Rebound if Yes in case of deadlocks rebound message using Sailor rebound mechanism, number of rebound can be specified via environment variable: ELASTICIO_REBOUND_LIMIT recommended value 3

Input fields description

image As an input metadata you will get a Primary Key field to provide the data inside as a clause value.

Execute stored procedure

This action calls stored procedure from selected DB Schema and Stored procedure name

Input fields description

  • DB Schema - a schema that contains a procedure to call. Must be selected from the dropdown list before Stored procedure name
  • Stored procedure - a name of a procedure to call, can be selected from the dropdown list

Metadata generates automatically using IN & IN OUT procedure parameters for input, and OUT & IN OUT procedure parameters for output.

As array fields this action now support ONLY:

  • CURSOR (as SQL type)
  • REF CURSOR (as ORACLE type) The result for this type of fields would be returned as an array of JSON objects.

This action DOES NOT processing MSSql @RETURN_VALUE.

  • For MySQL component same to DATABASE is same to SCHEMA by it's definition, so DB Schema dropdown is empty for MySQL.

  • MSSQL DB stored procedures has only IN and INOUT fields.

Usage case example

For Oracle DB procedure:

create PROCEDURE "INSERT_EMPLOYEE"(
        i_emp_id IN EMPLOYEE.EMPID%TYPE,
        i_name IN EMPLOYEE.EMPNAME%TYPE,
        i_department IN EMPLOYEE.DEPARTMENT%TYPE)
IS
BEGIN
  INSERT INTO EMPLOYEE (EMPID, EMPNAME, DEPARTMENT)
  VALUES (i_emp_id, i_name, i_department);
END;

Component generates next metadata:

image

Upsert Row By Primary Key

The action will execute SELECT command from a Tables dropdown field, as search criteria can be used only PRIMARY KEY, and execute INSERT command by PRIMARY KEY with specified field, if result does not found, else - action will execute UPDATE command by PRIMARY KEY with specified field. The action returns only one result row (a primary key is unique).

  1. Find and select jdbc-component in the component repository image

  2. Create new or select existing credentials image

  3. Select action "Upsert Row By Primary Key" from list image

  4. Select table from Table dropdown list image

  5. Specify input data (field with red asterisk is Primary key), and click "Continue" image

  6. Enable rebound mechanism if needed image

  7. Retrieving sample image

  8. Retrieve sample result image

  9. Click "Continue" image

  10. Finish component configuration image

Input fields description

  • Enable Rebound if Yes in case of deadlocks rebound message using Sailor rebound mechanism, number of rebound can be specified via environment variable: ELASTICIO_REBOUND_LIMIT recommended value 3 As an input metadata you will get all fields of selected table. PRIMARY KEY is required field (will mark as asterisk) and other input fields are optional. image

Create or update record (Deprecated)

This action exists in JDBC component only for backward compatibility. Please use Upsert row by primary key instead.

Select (Deprecated)

This action exists in JDBC component only for backward compatibility. Please use Select action instead.

Known limitations

  1. Only tables with one PRIMARY KEY is supported. You will see the message Table has not Primary Key. Should be one Primary Key , if the selected table doesn't have a primary key. Also, you will see the message Composite Primary Key is not supported , if the selected table has composite primary key.
  2. Only following versions of database types are supported:
  • MySQL - compatible with MySQL Server 5.5, 5.6, 5.7 and 8.0.
  • PostgreSQL - compatible with PostgreSQL 8.2 and higher
  • Oracle - compatible with Oracle Database 8.1.7 - 12.1.0.2
  • MSSQL - compatible with Microsoft SQL Server 2008 R2 and higher
  1. The current implementation of the action Upsert By Primary Key doesn't mark non-nullable fields as required fields at a dynamic metadata. In case of updating such fields with an empty value you will get SQL Exception Cannot insert the value NULL into.... You should manually fill in all non-nullable fields with previous data, if you want to update part of columns in a row, even if data in that fields doesn't change.
  2. The current implementation of the action Execute stored procedure doesn't support ResultSet MSSQL output.
  3. The current implementation of the action Execute stored procedure doesn't support any array types parameters. (MySQL does not have schemas by definition)
  4. Rebound mechanism only works for this SQL State:
  • MySQL: 40001, XA102
  • Oracle: 61000
  • MSSQL: 40001
  • PostgreSQL: 40P01
  1. If your database server configured to custom timezone (differ from UTC) JDBC driver may convert time appropriate - for example if you want to use Insert action with MySQL which is configured to +2:00 time zone and provide 2022-01-01 15:00:00 as value to some datetime field in database it will be saved as 2022-01-01 17:00:00.

License

Apache-2.0 © elastic.io GmbH