Package org.aksw.jena_sparql_api.io.endpoint
package org.aksw.jena_sparql_api.io.endpoint
-
ClassDescriptionClass for allowing an arbitrary number of concurrent reads to a file while it is being written to.A wrapper for a channel that blocks upon reaching the end of its channel When there is no more data, the reader can either waits for a notify on referenceChannel.Every destination must be usable as a supplier for input streams.Destination from an existing fileDestination of a file that does not yet existSwitching destination If a file for a DestinationFromFileCreation finishes, it can replace itself with a DestinationFromFile Right now I am not sure whether this is useful: If a stream is requested, it has to do all the preparation work anyway If we think of a source + filter as a virtual source, then it is actually ok: Let's not forget that a destination is a supplier for input streams Repeatedly executing such a workflow may re-use prior generated files instead of repeating the process again from the source file However, the purpose of this framework is not to do full caching, but only abstract java and system filters, but allow controlling what should happen with the files in case they get generated.An engine is a factory for executions that process input to output.A hot file is a currently running file creation to by another thread or process.An InputStreamSupplier is a source for InputStreams Obtaining an instance from the supplier should be near instant.