Skip to content

severinh/ethz-acd-global-type-inference

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Javali Compiler with Global Type Inference

Prerequisites

Make sure that you have installed JDK for Java 7. The project uses the build system Maven. On Ubuntu, install it using

sudo apt-get install maven

Maven will fetch all dependencies, generate the parser, compile the code and build the resulting JAR file. The dependencies include:

Building

Run the following to build the whole compiler:

mvn install

Running

To compile a single Javali source file 'foo.javali', run

./javalic foo.javali

To get the list of available options (type inference mode, optimization), run

./javalic --help

Testing

The Javali files for end-to-end tests are contained in src/test/javali. We reuse the reference compiler, but in cases where the output of the reference compiler deviates from the expected output of our compiler, we use overrides. This also applies to Javali files that use the extended syntax as described in the report (var keyword).

Overrides are represented as files with an override suffix. We run the tests for all Javali files using each implementation of type inference, and also without any type inference. To that end, we use different suffixes depending on what type of local inference the override applies to.

  • .override applies to any type inference implementation
  • .override.gti applies only to global type inference
  • .override.ltiwc applies to local type inference with constraints
  • .override.ltil applies to the efficient local type inference

Example:

  • arrays/InvariantArrayTest.javali.semantic.ref.override

    The first line is empty, indicating that our compiler with type inference should not yield a semantic error (in contrast to the reference compiler). The remaining lines usually contain comments explaining why the override exists.

About

Javali compiler with global type inference

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages