Github user zentol commented on the pull request: https://github.com/apache/flink/pull/1045#issuecomment-134203619 InvalidTypesException (from here on abbreviated as ITE) is no longer unchecked. For this to work i had to make changes in surprisingly many classes so let's break it down: * In general, InvalidProgramExceptions are now thrown for unrecoverable conditions, where previously an ITE would be thrown.. * TypeExtractor methods that return a TypeInformation will never throw an ITE and instead return a MissingTypeInfo if it is allowed, otherwise an InvalidProgramException. * One public TypeExtractor method now throws an ITE, which is getParameterType(). This method is heavily used in the TypeExtractor itself as well, so i didn't see a way to fix this in a non API breaking way that doesn't rely yet again on unchecked exceptions. * affects hadoop-compatibility functions, which now catch them * A few OperatorClasses catched ITE's in returns(TypeInformation ...) and then manually created a MissingTypeInfo. This now happens in the TypeExtractor directly. * TypeInformation classes threw an ITE in getInfoFor(Class ...) if the given class didn't match the TypeInformation class. These have been changed to IllegalArgumentExceptions. * DataSet/DataStream threw an ITE in getType() if the type was a MissingTypeInfo, changed to InvalidProgramException. * similiar issue in StreamExecutionEnvironment
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---