Closed joe-angell closed 4 years ago
@msettles @samhunter @dstreett I'm going to call this good for now unless you have any comments. It seems something is going on with superDeduper, it is outputting a different number of reads on the github regression than it is locally or on sam's machines. I don't know why that is yet, but I'll continue to look into that and address it in another pr.
Yay all fixed, @dstreett remember virtual functions must be virtual, and virtual classes need virtual destructors, ;P. I should probably look for some others that i missed.
simple regression script to run all the programs with default parameters and stores the output. When a pr is created, the regression test will run comparing the output, if the output has changed it will fail. Upon fail the user must update the regression data by running the regression script with the
regen
parameter, (after validating that the differences are good).