An approach to collect object graphs for data-structure live programming based on a language implementation framework

Data-structure live programming environments execute a program, collect object graphs (objects and their mutual references) created and modified during the execution, and visualize the graphs as a node-link diagram. Existing implementations collect object graphs by instrumenting checkpoints, which traverse reachable objects, into every necessary points in the program. Since the cost of each checkpoint is proportional to the number of existing objects, its overhead can be huge. This paper proposes (1) a method to collect object graphs by recording object creation and modification events into an efficient data structure, and (2) to implement the object graph collection mechanism by extending a language implemented on top of a language implementation framework. As a result, the overhead of object graph collection is merely proportional to the number of object creation/modification operations in total. We implemented the proposed mechanism for the Kanon data-structure live programming environment by extending GraalJS, which is a JavaScript implementation on the Graal/Truffle language implementation framework. We compared our new implementation against the original Kanon, which is based on checkpointing, and confirmed that our implementation improves program execution (and data collection) speed, and has sufficiently small overheads to reconstruct object graphs.