This is the public repo for the project:
An Empirical Analysis of the Transition from Python 2 to Python 3
Here you will find:
getting-the-corpus: details on how to get the Python programs we analysed as part of the study (we don't copy the corpus here), including the GitHub URL and the SHA ids for the relevant versions.
analysis-code: The Python and shell scripts that produce the data that we discuss in the paper; see the README file here for details on how to reproduce these tables. The code here will build working versions of PyComply for the various releases in the Python 2 and Python 3 series.
archived-data: A copy of the data we generated using our code; most notably the pass rates for the various applications and various versions of PyComply.
esem-paper-apr-2017: this is just here for the record, and is the status of the project when we submitted the paper to the ESEM 2017 conference.