SugarJ Eclipse plugin (recommended)
Visit the SugarJ web site http://sugarj.org
Installation
- Install Eclipse (follow instructions on eclipse.org).
- Start Eclipse.
- In Eclipse, select 'Install New Software' in the 'Help' menu.
- In the 'work with' field, copy the SugarJ update site
http://update.sugarj.org and hit enter.
- Be sure to deselect the 'Group items by category' checkbox on the bottom of
the window.
- Select the latest instance of Sugarclipse and click continue. This will
install the SugarJ compiler, Spoofax and the Sugarclipse plugin.
In addition, please ensure enough stack space (about 4-16 MB) is available for
the SDF parser. You can set the stack space of your Java runtime using the
-Xss16m command line argument when starting Eclipse or setting -Xss16m in your
eclipse.ini file.
Setting up a SugarJ project
- Create a new Java project.
- As for now, we need to register the SugarJ builder for this project by hand:
Open your project's '.project' file in any text editor and replace the Java
build command by the following code:
org.sugarj.editor.SugarJBuilder
- We're ready to go. Note: SugarJ source files must have the file extension
".sugj".
SugarJ standalone compiler
Installing SugarJ
The SugarJ compiler is almost self-contained and only requires an installation
of a Java runtime version 6 or higher. Download sugarj.zip
and
extract it to a location of your choice. The directory structure of
the archive is as follows.
sugarj/
bin/ Scripts to invoke SugarJ
sugarj Compiler for *nix
sugarj.bat Compiler for Windows
sugh Alias of `sugarj -l haskell` for *nix
sugh.bat Alias of `sugarj -l haskell` for Windows
sugj Alias of `sugarj -l java` for *nix
sugj.bat Alias of `sugarj -l java` for Windows
case-studies/ Sample SugarJ projects
lib/ The back end
README.md This file
Adding sugarj/bin
to the PATH
environmental variable of your
platform will allow invocation of SugarJ scripts without prefixing
them by a location.
Invoking SugarJ
If the current directory is your working directory containing the source files,
calling SugarJ is very easy:
bin/sugarj -l java closures/Test.sugj
The -l
flag is necessary to specify your host language of choice. Currently,
we support Haskell, Java, and Prolog.
If your source files are located in case-studies/closure/src
, you can invoke
the compiler like this on *nix:
bin/sugarj -l java \
--sourcepath case-studies/closures/src \
-d case-studies/closures/bin \
concretesyntax/Test.sugj # file(s) to compile relative
# to sourcepath
On Windows:
bin\sugarj -l java ^
--sourcepath case-studies/closures/src ^
-d case-studies/closures/bin ^
concretesyntax\Test.sugj
The generated Test.class
may be executed thus:
java -cp case-studies/closures/bin concretesyntax.Test
Compiler options
--atomic-imports Parse all import statements simultaneously.
--cache <arg> Specifiy a directory for caching.
--cache-info Show where files are cached
-cp,--buildpath <arg> Specify where to find compiled files. Multiple
paths can be given separated by ':'.
-d <arg> Specify where to place compiled files
--full-command-line Show all arguments to subprocesses
--gen-files Generate files?
--help Print this synopsis of options
-l,--language <arg> Specify a language library to activate.
--no-checking Do not check resulting SDF and Stratego files.
--read-only-cache Specify the cache to be read-only.
--silent-execution Try to be silent
--sourcepath <arg> Specify where to find source files. Multiple
paths can be given separated by ':'.
--sub-silent-execution Do not display output of subprocesses
-v,--verbose Show verbose output
--write-only-cache Specify the cache to be write-only.