One of the benchmarks that SuanShu uses is R. So, the results of SuanShu and R should match.
What are the differences that you see? Could you post the R code (with data and output) so that we can take a look at it please?
BTW, I might have mis-read your comment
runs a regression between Y and X (assume 2 time series data)
This is not Johansen’s method. This is more similar to the Engle and Granger’s method. They in general do not produce the same outputs.
SuanShu uses the Johansen’s maximum likelihood method.