Automated software transplantation

Abstract

Automated transplantation would open many exciting avenues for software development: suppose we could autotransplant code from one system into another, entirely unrelated, system. This paper introduces a theory, an algorithm, and a tool that achieve this. Leveraging lightweight annotation, program analysis identifies an organ (interesting behavior to transplant); testing validates that the organ exhibits the desired behavior during its extraction and after its implantation into a host. While we do not claim automated transplantation is now a solved problem, our results are encouraging: we report that in 12 of 15 experiments, involving 5 donors and 3 hosts (all popular real-world systems), we successfully autotransplanted new functionality and passed all regression tests. Autotransplantation is also already useful: in 26 hours computation time we successfully autotransplanted the H.264 video encoding functionality from the x264 system to the VLC media player; compare this to upgrading x264 within VLC, a task that we estimate, from VLC's version history, took human programmers an average of 20 days of elapsed, as opposed to dedicated, time.

Automated Transplantation of Call Graph and Layout Features into Kate

“…We presented a framework for transplanting a feature between two unrelated systems and a tool to implement our approach [1] in our recent ISSTA paper [2], so we provide merely a summary here to make the paper self-contained. Given a host H program that is lacking a feature of interest, a donor program D, that implemented the feature, and a lightweight annotation system, our tool, µ SCALPEL, attempts to autotransplant the feature from D into H. The feature of interest in the donor is called the Organ.…”

“…GP explores combinations of statements on the vein, and in the host-donor variable mappings, that will enable the organ to execute in the host environment, guided by testing. The first stage of our approach uses context insensitive slicing on the call graph of the donor program to construct a map, with the key being the variables available in the vein, and the values being the variables in scope at the implantation point in the host [2]. GP is used to transplant the feature in the host system, having the host-donor map, and the code base represented by these context insensitive slices.…”

“…The search space has two dimensions: the variable mapping and the statements available to form the transplant itself. The tool that implements out approach, µ SCALPEL, is publicly available [2].…”

“…We recently introduced a search based technique for automated software transplantation [2,7]. Guided by dependence analysis and testing, our approach uses a variant of genetic programming to identify and extract useful functionality from a donor program, and transplant it into a (possibly unrelated) host program.…”

“…Guided by dependence analysis and testing, our approach uses a variant of genetic programming to identify and extract useful functionality from a donor program, and transplant it into a (possibly unrelated) host program. We implemented our autotransplantation approach as a tool called µ SCALPEL, which is publicly available [1].…”