Comment on page
5.8 Release Notes
Cinchy version 5.8 was released on October 31, 2023
Cinchy version 5.8.1 was released on November 30, 2023
For instructions on how to upgrade your platform to the latest version, please review the documentation here.
Select the links below to see the relevant upgrade guide and release notes.
The following section contains patch notes for the 5.8.1 release.
Alphabetical sorting of queries in dropdowns
We implemented alphabetical sorting for queries in the Connections listener UI RunQuery and Cinchy Query dropdowns. This streamlines navigation and simplifies query selection for users.
- We fixed an issue that ensures that Destination mappings in dropdowns now display the alias, instead of the original column name.
- We resolved an issue where dropdowns were not correctly loading data due to user permissions on system tables. This fix, involving an API change, ensures that dropdown data reflects appropriate user access levels.
- We resolved an issue where the Query dropdown was not populating when you selected RunQuery in Connections listener UI.
- We resolved a rendering issue in the Connections listener UI, where line breaks in the topic JSON were causing display problems.
Cinchy now supports a new Cinchy event-triggered source: SOAP API. This new feature initiates a SOAP call based on Change Data Capture (CDC) events occurring in Cinchy. The SOAP response then serves as the source for the sync and can be mapped to any destination.
The POST endpoint for Saved Queries now automatically serializes hierarchical JSON to text when the content-type is
application/json. This update now supports values that are objects or arrays. This eliminates the need for manual serialization and makes it easier for developers to work with Saved Queries.You can now pull specific data from REST API response headers using .NET regex capture groups. This feature gives you more control and flexibility in collecting the data you need when using REST API destinations. For more information, see the REST API destination page.
Extract variables from response header
We improved the user experience for header row settings for delimited files. The following improvements have been added.
- Use Header Row Checkbox: Controls visibility of column names and Load Metadata button.
- Schema Columns Warning: Informs users about column order when header row is disabled.
- Modal Warning: Explains schema column reset when disabling header row.
- Header Record Row Number: Specifies row to use as header.
We added links next to Cinchy Tables referenced in the UI. These links directly open the respective table, making navigation more seamless.
You can now use drop-down menus for selecting Cinchy tables and queries for both Cinchy sources and destinations. This feature replaces the previous method, where users had to manually type in their selections.
We added a warning to the Schema sections of multiple Sources to mitigate issues due to mismatched column order. This warns users that the column order in the schema must match the source/destination configuration. The changes affect the following data sources:
- LDAP
- Excel
- Binary
- Fixed Width
- Cinchy Query
We added a Listener section to the MongoDB Collection (Cinchy Event Triggered) and REST API (Cinchy Event Triggered) Sources. You can now manage the event trigger within the Connections UI. This reduces the complexity of managing the Listener Config table.
We improved the implementation of
DataPollingConcurrencyIndex. We also added additional logging in the Data Polling Listener to enhance monitoring.When configuring a connection source with text columns, it's possible to specify a JSON content type. This instructs the system to interpret the contents as a JSON object and pass it through as such. This is useful when the target (such as Kafka) supports and expects a JSON object for a specific target column. When setting this option, the value should always be valid JSON.
Alternatively, the original, default behaviour of treating text columns as plaintext is unchanged. As plaintext, the contents of the column will be passed through as a string, even if it could be interpreted as JSON.
- We modified the Data Polling mechanism to enhance the reliability of message delivery to Kafka.
- We fixed a bug in CQL on PostgreSQL that caused the
DATEADDfunction to truncate input dates down toDAYprecision. Now, you can expect more accurate date manipulations without losing finer time details. - We fixed an issue where the UI failed to display Batch Data Sync results and instead showed a generic exception message. The jobs tab in the UI now opens without any API failure appearing in the browser's network console.
- We resolved an issue that caused large batch delta syncs to fail in Cinchy.
- We fixed an issue where Cinchy CDC Delete events weren't sent to the destination using Delta. For example, Deletes and Approved Deletes now successfully insert records into Kafka when deleted from a Cinchy table.
- We fixed the issue of concurrent updates failing due to a Primary Key (PK) violation on the History table by adding a retry mechanism. This fix aims to make Cinchy more robust when making concurrent updates.
- We resolved an issue during where the Cinchy destination would still be queried during a delta sync.
- We fixed an issue with data syncs that would fail on executed queries that returned large numbers of records on Cinchy Table destinations.
- We resolved issues with parameter resolution in Rest API authentication requests.
- We resolved an issue where the Load Metadata button was failing to connect to DB2 databases when fetching schema information.
- We fixed an issue where the Mapping UI would disappear in the Destination Section for Cinchy Event Broker to MongoDB Collection syncs, where Sync Actions were set to Delta.
- We fixed an issue where system columns like Created By, Created, Modified By, Modified, Deleted, and Deleted By weren't appearing in the topic columns dropdown in the Listener UI.
- We fixed a bug where the model loader failed to update when you added a description to a calculated column. The table now saves correctly when making changes to calculated columns.
- We fixed an issue that prevented table selection from the drop-down in Cinchy Event Broker's listener configuration.
- We resolved an issue where the
Lookup()function in the Filter field for Cinchy Tables wasn't behaving as expected. - We restored the default timeout setting for
HttpClientto over 100 seconds.
Last modified 7d ago