Linux environment, sbt, python3.
I test the script in wsl Ubuntu, sbt 1.6.1 and python 3.8. The package regex and pandas are required. In the python script, I use regex instead of re to do the pattern matching, since re doesn't support the variable width of lookahead regex.
Run the command
./autotest --scala SCALA-VERSION --scala-native SCALA-NATIVE-VERSION --benchmark-list <benchmarks list or non to run all>
Or use the default configuration
./autotest
The script automatically test the correctness and performance of incremental compilation in Scala Native. After running this script, the performance is presented in the file result. There are some keywords in this file:
-
WARMUP PHASE: the phase is used for warmming up sbt. Just ignore it.
-
COMPILE PHASE: After cleaning the
target, this phase compiles the scala project from scratch. -
EDIT PHASE: We modify some source codes of scala project and then recompile it. In this phase, only very little source code is changed.
-
TRANSITION PHASE: We directly change the source code from one main class to another main class. Therefore many libraries changes and source code changes are applied. Since too many changes are applied, the advantages of incremental compilation are suppressed. Therefore, we only use it to test the correctness of incremental compilation.
The result is presented in the comparison directory.