This repo contains a minimal skeleton to run semantic analysis on a corpus of 2 million lines of Scala code. A breakdown of which projects are included in the corpus and how many lines of code each project contains can be found here https://docs.google.com/spreadsheets/d/1btkCiF30Wb9MJti6LDc9og788XqXBKgwEhIdKb9aloc/edit#gid=1799045603
Analyses run fast since we don't have to re-compile the corpus to run a new analysis. Simple analyses on the full corpus run in 5 seconds on my laptop, for example to count the number of call-sites to a particular symbol. Larger analyses that require parsing can take ~60 seconds, YMMV.
To run analyses locally, first download https://drive.google.com/open?id=0B5BBsTPBxGcMYnlXSk5KUk40Q0k
cd scala-experiments
unzip -d target/v13 path/to/semanticdb.v13.zip
sbt run
A report is printed out to console and written to markdown file under
target/report-<TIMESTAMP>.md.