--- title: "18th of May 2026" slug: "18th-of-may-2026" updated: 2026-05-20T10:45:47Z published: 2026-05-20T10:45:47Z canonical: "knowledge.curiositysoftware.ie/18th-of-may-2026" --- > ## Documentation Index > Fetch the complete documentation index at: https://knowledge.curiositysoftware.ie/llms.txt > Use this file to discover all available pages before exploring further. # 18th of May 2026 # New Features and Enhancements - Added **SSO support for Cloud deployments**. SSO was previously available for on-premise customers only and is now fully supported in Cloud environments. - Introduced **Rule Set Parameterisation of API Calls**, allowing API calls within rule sets to be parameterised for more flexible and reusable data generation configurations. - Added **OAuth and API Proxy support**, enabling integration with external services using OAuth authentication via an API proxy. - **Optimizer improvements**: Reworked the optimizer with a focus on constraints performance, including a fix to all-pairs logic. Rules can now be dynamically resolved, improving the handling of complex constraint scenarios. - Added the ability to **set a custom AI connector as the global default**, making it easier to standardise AI usage across the platform. - Added **horizontal scroll buttons on lists** for improved navigation of wide list content. - Added support for **Fixed Width file schema** in the front end. - Introduced the ability to **define custom labels for specific node types**, such as Start or End nodes, for clearer model readability. - When **linking subflows**, only end nodes associated to the respective start node are now displayed, reducing noise and preventing incorrect connections. - Added **Favourite Folder and Subfolder support**, allowing users to mark folders and subfolders as favourites for faster navigation and a better save experience. - Added the ability to **configure custom fields at the Test Path level**, supporting more granular tracking and reporting per path. - Added the **trim() function to masking where clauses**, expanding the available functions for masking rule conditions. - **Definition Version Scan** now includes a select box to add items to view, along with a new option to add directly to a tag. - **Modeller variables in Data Painter** now show model names on hover, making it easier to identify the source of a variable at a glance. - Improved the **Preview Data screens** with a cleaner, more usable UI. - Added **Key Vault support in the TDM API**, enabling secure retrieval of credentials and secrets from a Key Vault. - Added support for **encrypted passwords in configuration** across Windows VM, Linux VM, and other deployment types. - Introduced **cascading of Gherkin tags when tagging a subflow**, so tags applied to a subflow propagate correctly through the model. - Improved management of large-scale data models with better **searchability and grouping** in the Definition Version Scan. - Improved **logging for uninitialised variables**, making it clearer which variable is out of range when an index error occurs. - Updated **GoJS to version 3.1.8** and **Spring Boot to 4.0**. - Stopped using **Zookeeper for service discovery**, as part of ongoing infrastructure improvements. --- ## Bug Fixes - Fixed an issue where **test path names reverted** to their original value after being modified. - Resolved a problem where **search returned zero results** when searching for models in the Project tab. - Fixed a bug where the **AI Assistant button appeared regardless of licence**, now correctly controlled by licence entitlement. - Fixed an issue where **models could not be moved** from one project to another. - Corrected a problem where the **submit job button** never reported completion on the same screen. - Fixed the **Model tab reloading to the homepage** when a user session expired and they signed back in. - Resolved an issue where **test paths continued to show as invalid** after being repaired. - Fixed intermittent **removal of tags** from assets. - Fixed an issue where **updating the UI type in a server process** did not take effect. - Resolved a bug where **duplicate pipeline runs** were triggered when regenerating a module collection in Quality Modeller. - Fixed an issue where **regenerating code in a linked module** after adding a custom function caused saved element locators in the Page Object to be lost. - Fixed an error when **moving functions between collections**. - Fixed the **Run DDL button missing** on CSV file types. - Corrected an issue where **XSD imports with matching elements** caused a scan save error even when the XSD was different. - Fixed a bug where **masking function edits** could fail when removing a parameter that had been used in a rule set. - Resolved an issue where **test data overrides** did not apply correctly when the same model was reused multiple times within a master model. - Fixed a **Paypal v3 Swagger creation issue**, along with a console error and hang when skipping Swagger creation. - Fixed an issue where **refreshing the run spec** for a data activity caused assigned values set to a variable or expression to revert. - Fixed handling of **odd characters in CSV column names** during CSV to DB import. - Resolved a **non-execute DB query connection error**. - Fixed inconsistencies in the **AI data generator rule prompt** across environments. - Cleaned up **AI-generated rules** to remove unnecessary or malformed output. - Fixed an error when **attaching a file template** in the Message Management activity. - Resolved an issue where **upgrading a data generation rule set** deleted the previous rule set version. - Fixed an issue where **importing a subflow** prevented linking to another node. - Corrected **child table ordering** in data generation scenarios. - Fixed a bug where **C drive storage filled up** with logs over time. - Resolved an issue with **opening imported ARD models** causing Quality Modeller to hang. - Fixed the **Subset rule set** to correctly disable foreign keys added to a definition. - Corrected a **Unique Condition inconsistency** between the API and VIP sides. - Fixed a **CData connection** handling issue. - Resolved a problem where **pipeline data generation** did not correctly assign the parCuriosityRunID value.