pfff

pfff is a set of tools and APIs to perform some static analysis, dynamic analysis, code visualizations, code navigations, or style-preserving source-to-source transformations such as refactorings on source code. For now the effort is focused on PHP but there is preliminary support for Javascript, C++, Erlang and other languages. There is also preliminary support for OCaml code so that the framework can be used on the code of pfff itself.

For each languages there are mainly 2 libraries, for instance parsing_php.cma and analysis_php.cma, that you can embed in your own application if you need to process PHP code. See the demos/ directory for example of use of the pfff API. See also docs/manual/Parsing_xxx.pdf and docs/manual/Analyzis_xxx.pdf for more documentation on how to use or extend pfff.

pfff is also made of few tools:

• pfff, which allows to test the different parsers on a single file
• pfff_db, which does some global analysis on a set of source files and store the data in a marshalled form in a file somewhere (e.g. /tmp/db.json)
• pfff_db_heavy, which does some heavy language specific global analysis on a set of source files and store the data in a database somewhere (e.g. /tmp/pfff_db/)
• codemap, which is a gtk and cairo based source code visualizer/navigator/searcher leveraging the information computed previously by pfff_db
• sgrep, a syntactical grep
• spatch, a syntactical patch
• scheck, a bug finder
• scheck_heavy, a bug finder leveraging the information computed previously by pfff_db_heavy
• stags, an Emacs tag generator

For more information, look at the pfff wiki here: http://github.com/facebook/pfff/wiki/Main as well as the docs/manual/ directory.

Usage for pfff:

$ ./pfff -parse_php demos/foo.php

or

$ ./pfff -dump_php demos/foo.php

You can also look at ./pfff --help

Usage for pfff_db:

$ ./pfff_db -lang ml -o /tmp/pfff.json ~/pfff

to analyze all the .ml and .mli files under ~/pfff and store metadata information (the database) in /tmp/pfff.json

Usage for codemap:

$ ./codemap -with_info /tmp/pfff.json ~/pfff

This should launch a gtk-based GUI that allows you to visualize source code and perform some code search.

Usage for pfff_db_heavy:

For now only PHP is supported for the heavy analysis. To build the heavy database do:

$ ./pfff_db_heavy -metapath /tmp/pfff_db/ ~/www/

to analyze all the .php and .phpt files under ~/www and store metadata information (the database) in /tmp/pfff_db/. It may takes some time, for instance 20 minutes for a PHP codebase with many million lines of code.

Once this is done you can use some of the flags of pfff_db_heavy to do some analysis as in:

$ ./pfff_db_heavy -deadcode_analysis /tmp/pfff_db/

More information

Look at the pfff wiki here