Skip to main content

​
v4.0.0 - 2026-11-06

Kodda 4.0 brings a refreshed interface, API keys for integrations, SQL history for mappings and a stronger, up-to-date platform underneath. Screens look different, but mappings, workflows, schedules and connections work the same way they did in 3.x.

​
Highlights

  • Application API keys — administrators can issue and revoke long-lived keys for scripts and integrations.
  • SQL history for mappings — every change to a mapping’s SQL is recorded and can be compared side by side.
  • New light theme — light mode is the default, and every screen has a refreshed, consistent look.
  • Mapping trees organized by target — the Map and Workflow screens group mappings by where their data lands.
  • Stronger security — better protection against common web attacks and up-to-date security fixes throughout.

​
New Features

  • Application API keys — Admins can create API applications and generate keys for scripts and integrations from the new API Keys page under the Manage menu. Each key is shown once when created, shows when it was last used, and can be revoked individually or by deactivating the application. API keys replace the login tokens used by the 3.x API; see Breaking changes below. See API Keys.
  • SQL history and compare for mappings — Every change to a mapping’s source SQL is recorded with who made it and when. Open History from the Source SQL panel to browse the history and compare any two versions side by side. History is kept when a mapping is re-imported. See Source SQL History.
  • Light theme — A new light theme is the default for users who have not chosen one. Dark mode remains available from the sun and moon icon in the top navigation bar. See User Interface.
  • Workflow screen search and canvas — The Design Workflow screen has a search box for finding components in a workflow, and the canvas fits the whole workflow in view when opened.
  • Faster upgrades for the browser — JavaScript and stylesheet files are versioned by content, so after an upgrade browsers always load the new files without a manual cache clear.

​
Security Improvements

These changes apply automatically and need nothing from you.
  • Stronger protection against common web attacks.
  • API keys are stored securely and shown only once, at creation.
  • Improved logging hygiene by reducing unnecessary detail in log output.
  • Outdated and unused components were removed.
  • The platform and its database drivers are on current versions with the latest security fixes.
  • Old sessions and job history are cleaned up automatically.

​
Improvements and Fixes

​
Interface

  • Every screen has a refreshed icon set and consistent labels, buttons, dialogs and table layouts.
  • Trees in Map, Workflow and Scheduler use consistent colors, highlight the selected item and scroll to it automatically.
  • Target table property labels are shorter and clearer.

​
Mappings and Execution

  • A mapping whose source folder has no matching files completes with a warning instead of failing.
  • Delimited file targets handle mixed-timezone timestamps correctly.
  • The file profiler in the Upload Files screen is faster and handles large numeric columns.
  • Uploaded CSV files treat empty strings as NULL and trim surrounding whitespace.
  • Run logs are quieter and show the load type by name.

​
Connectors

  • Oracle — uppercase catalog names and primary-key columns are matched correctly; national character types map to NCHAR, NVARCHAR2 and NCLOB.
  • SQL Server — Azure SQL databases no longer fail a run when version checks or trace flag setup are not permitted.
  • Snowflake, MySQL and Redshift — assorted fixes, and a longer discovery timeout for Redshift serverless connections.
  • Database drivers for Oracle, SQL Server, Snowflake and Redshift are updated.

​
REST API

  • Malformed import requests return a clear error message instead of a server error.
  • Requests for a map or workflow that does not exist return a not-found response.
  • Pagination for map and workflow execution history is correct.

​
Breaking Changes

  • API authentication — The 3.x login endpoint (/kodda/api/v1/login) and its JWT access tokens are removed. Scripts and integrations that call the REST API must switch to an application API key, issued by an administrator from the API Keys page and sent as Authorization: Bearer <key>.
  • Generate Mapping — The All Schemas and Tables generation option was removed.
  • File targets — The DDL option on delimited file targets is hidden.

​
Behavior Changes

  • The mapping trees on the Edit Mapping and Design Workflow screens are now grouped by target connection, then target schema, then target table. In 3.x they had an additional grouping of map type, source connection and source schema. Use the tree search box to find a mapping by name.
  • A mapping run that finds no source files now succeeds with a warning rather than failing. Alerts that relied on the failure should be reviewed.
  • Property names on target tables and loads were renamed for clarity; the underlying settings and their values are unchanged. See Target Properties.
  • The default theme is light. Your theme choice is remembered per browser.
  • Run logs are less verbose by default.

​
v3.2.2 - 2026-04-08

​
New Features

  • Added a method to test connectivity for file-based connections

​
Updates

  • File names are now logged less frequently when processing delimited files, reducing noise in log output.
  • Previously, mappings using a file source would fail if no files were found. This has been updated to issue a warning instead, allowing the mapping to complete successfully with 0 rows processed.
  • The Redshift ODBC Driver is no longer used. This reduces external dependencies and simplifies installation requirements.
  • The Oracle driver has been updated to the latest version.

​
v3.2.1 - 2026-01-15

​
New Features

​
Parquet File Support

  • Added native support for Parquet files - You can now directly process Parquet files as data sources in your pipelines
  • SQL-like filtering for Parquet files - Filter Parquet file data using familiar WHERE clause syntax, enabling more efficient data processing by reading only the records you need

​
Enhanced Load Types

  • Delete Load Type - New load type that enables deletion of records in target tables via Kodda mappings
  • Incremental Delete Load Type - Perform incremental deletions in target tables

​
Performance Optimizations

  • Oracle Direct Path Loading - The Oracle data loader now automatically uses direct path load when data types are compatible, significantly improving load performance for large datasets
  • Bypass Work Tables for Full Refresh - Added a new load parameter for supported database targets that allows skipping the work table stage during full refresh loads, reducing overhead and improving performance

​
Improvements

Upgraded to Python 3.12 - The platform now runs on Python 3.12, providing improved performance and access to the latest Python features Performance improvements and bug fixes - Multiple enhancements across the platform to improve performance, resolve outstanding bugs

​
Removed Connectors

The following data source connectors have been removed in this release:
  • Google Analytics 3 - Google Analytics 3 has been sunset by Google.
  • Apache Hive - Support for Hive has been discontinued. platforms
  • IBM Netezza - Netezza connector support has been removed

​
V3.1.6 - 2025-09-16

  • Improvements to service logging
  • Bug Fixes
    • Resolves a bug where timestamp incremental values from files were being formatted prior to storing.

​
v3.1.5 - 2025-07-24

  • Bug Fixes
    • Resolves a bug that prevents files from being read from the root of an S3 bucket

​
v3.1.4 - 2025-03-04

  • Bug Fixes
    • Resolves a bug that was preventing read only users from using the search bar
    • Set timeout interval to 60 seconds for redshift connections

​
v3.1.3 - 2024-10-22

  • New Features
    • Allow a sort property to be specified when selecting data from a database and writing to a file.

​
v3.1.2 - 2024-05-27

  • New Features
    • Objects with errors are placed at the top of the list when emailing workflow status.
    • Provide button to navigate back to SQL when generating a custom sql mapping with a syntax error.
    • Add search screen to home page
    • Add most recently updated maps by all users to home page
  • Bug Fixes
    • Fixed bug in browse data screen that prevented scrolling when expanding object tree.
    • Fixed bug that prevented connections used in email widgets from importing properly.

​
v3.1.1 - 2023-07-01

  • New Features:
    • Google Analytics 4 Connector: A new connector has been added, enabling data retrieval from Google Analytics 4.
    • Change Tracking Support for Sql Server: The platform now supports change tracking when sourcing data from Sql Server.
    • Handling of Empty Delimited Files: Delimited files that are empty will no longer cause an error and will be skipped instead.
    • “Create Empty File” Option: When generating a mapping that writes a delimited file, the new file target option “Create Empty File” is now available. By default, this option is disabled.
    • Incremental Value: A new feature allows users to prevent the incremental value from changing in a mapping if the workflow fails.
    • Reset Incremental Values at Runtime: Users now have the capability to reset incremental values for all mappings in a workflow during runtime.
    • Updated Landing Page: The landing page has been updated to display the last 10 mappings that the user has worked on.
  • Bug Fixes:
    • Improved Performance for Historical Log Files: Performance has been enhanced when fetching historical log files.
    • Scheduling UI Improvement: The scheduling UI no longer re-enables a paused job upon saving.

​
v3.0.2 - 2022-08-28

  • Changes
    • Migration of Kodda Application to Python 3.8
    • General Performance improvements
      • Workflows
        • Limitation removed for nesting of workflows
          • Data sets can be sent in excel format from a workflow
          • Custom From email address can be set when sending data sets from a workflow
          • Connections
            • Support added for Azure SQL pools
          • Metadata Repository
            • Updates to metadata repository to improve application performance.

​
v2.3 - 2021-08-27

  • Changes
    • Kodda Scheduler will no longer allow multiple instances of a scheduled job to run at the same time. (Scheduled jobs must be resaved to take advantage of this)
    • Kodda Restful API’s.
      • Connections
        • endpoint added to retrieve all available connections
      • Workflows
        • endpoint to retrieve all workflows
        • endpoint to retrieve workflow details
        • endpoint to retrieve workflow execution history
        • endpoint to run a workflow
        • endpoint to export a workflow
      • Maps
        • endpoint to retrieve all maps
        • endpoint to retrieve map details
        • endpoint to retrieve map execution history
        • endpoint to run a mapping
        • endpoint to export a mapping
      • Scheduler
        • endpoint to retrieve scheduler status
        • endpoint to retrieve all scheduled jobs
      • Tasks
        • Endpoint to retrieve a tasks status
      • Import
        • Endpoint to parse and list Import Objects
        • Endpoint to import objects
    • New Kodda function added for sql server extracts that leverage a binary value for incremental loads. When used Kodda will cast the stored value to a varbinary data type.
  • Bug Fixes
    • Fix bug related to parsing queries with optionally defined columns.
    • Fixed parsing of SQL Server queries that contain an OPTION clause. Kodda will now properly rewrite these queries to ensure the option is at the end of the query.
    • Fixed a bug that prevented applying changes on file upload screen.
    • Ensure queries are not split when push-down optimization is enabled and multiple extractors are defined.

​
v2.2 - 2021-06-06

  • Changes
    • Read Only - A new user group has been added to support read only users who wish to view Kodda mappings and workflows.
    • Emailing of Data Sets - Providers users with the ability to attach a dataset to an email task in a workflow.
    • Performance improvements and improved memory management when using Redshift as a data source.
    • Improved loading to postgres targets.
    • PGBouncer - When PGBouncer is enabled logs will be written to the syslog.
  • Bug fixes
    • Fixed bug when user selects to load Null Values as an empty string and column is part of a primary key.

​
v 2.1.11 - 2021-03-08

Changes
  • Reduce number of API calls required to identify available files in S3
  • Performance improvements when loading and generating mappings from S3 delimited files.
  • Addition of hotkeys to workflow screen to improve UX. (Alt+X to disable/enable components in a workflow)
  • Support added for multiple LDAP servers in setup.cfg
  • New mapping run-time parameter to enable failing a map when 0 records are read from the source.
  • Auto-refresh map and workflow operation screens
  • Allow viewing connection table and testing connections for non admin users.

​
v 2.1.10 - 2020-12-07

  • Bug fixes
    • Update pgbouncer default config to allow SSL connections
    • Bug Fix to object export - Allow unicode characters to be exported from table and column comments.
    • Bug Fix to Workflow engine - Fix bug for workflows with multiple dependency paths and a parent task in one path fails, while the second path is still running.

​
v 2.1.9 - 2020-11-14

  • Updates to workflow screen
    • Workflow audit history attributes have been added to east pane.
    • Added highlighting of objects on mouse over to better differentiate the mouse being over a workflow widget vs the workflow canvas.
    • New click through capabilities on workflow items. (I.e. click to drill to map or nested workflows).
    • Update to email notifications. Disabled tasks and objects are now included in workflow status emails to improve visibility into steps that were disabled at runtime.
  • Mappings
    • Reduced the number of database calls when generating or updating a mapping.
    • Set default file size when delivering data to Redshift and Snowflake. This will reduce reliance on developers having to set this parameter.
  • Inclusion of PG_BOUNCER:
    • PG_Bouncer is now included as part of the default build.This optional component can be enabled by adding the parameter pg_bouncer = enabled to the MetaDB and secDB sections of the setup.cfg file.