Flutter SDK 3.8.0

Features

• Upgraded Flutter SDK to use iOS SDK 4.27.3 / Android SDK 4.26.3 / JS/TS SDK 4.24.5
• Improved the performance of the getDataFile method. It now returns a cached DataFile instance that is refreshed whenever the SDK configuration is updated.
• Added a new dateModified property to the DataFile object returned by the getDataFile method.
• The getFeatureList method has been deprecated in favor of getDataFile().
• Added support for *during the current visit- and *during any of the last visits- settings across the following targeting conditions:
  • Converted Goal
  • Feature Flag
  • Web Experiment
  • Personalization
  • Exclusive Campaign
• Increased the default and minimum values of the activityTrackingIntervalMilliseconds parameter from 15 000 ms to 60 000 ms. This reduces the number of activity requests sent to the Data API and helps prevent 429 (Too Many Requests) responses. (Only iOS / Android)
• Tracking requests are now paused while the application is in the background (iOS / Android) or while the visitor is not genuinely interacting with the page (Web). This reduces the number of requests sent to Kameleoon servers and prevents inflated visit counts. To continue sending data in these cases, explicitly call flush(instant: true).
• Added support for variation simulation when feature flags are **inactive (OFF state)*- in evaluation methods such as getVariation, getVariations and isFeatureActive. (Only Web)
• flush(instant: true) now awaits the completion of the tracking request. (Only Web)
• The bundled JS SDK now uses the *full- build, which retains INFO and DEBUG log messages that were removed from the standard JS SDK bundle in version 4.24.0, so all KameleoonLogger log levels keep working. (Only Web)

Bug fixes

• Fixed an issue where, in certain scenarios, the SDK data file (configuration) was not updated at the configured refreshIntervalMinutes interval in polling mode and was instead refreshed only after the application restarted. (Only iOS / Android)
• Fixed an issue where the *Web Experiment- targeting condition could return incorrect non-targeted results. (Only Web)