Welcome to CUP2 beta

This is a development version of the upcoming new CUP release! It will include

• LR(0)
• canonical LR(1)
• LALR(1)
• Pager-LR(1)

and a few other parsing kernels. It will feature a pure-Java syntax and supports a modular object oriented architecture, advanced error handling, symbol region information and a more sophisticated ANT-task, and last but not least: Finally Unit-Tests for the components.

The most recent documentation document can be found  here

You can obtain the recent sources via

svn checkout https://www2.in.tum.de/repos/cup2/trunk

Current status: We managed to implement

• LR-Table-Parsing-Driver
• LR(0)/LR(1)/LALR(1) engines
• Serialisiation of parsetable
• JFlex-CUP2 patch
• Productive and unreachable nonterminal optimisation
• Grammar-Specification API
• ANT-task
• Token completion
• preliminary documentation
• random word generator
• Implemented context propagation links
• Automated conflict examples
• Parallelized automaton generation algorithms via threadpool
• specialised algorithm for First/Follow?-Sets based on strongly connected components

Still to do:

• feed JFlex-patch back
• CUP1toCUP2
• typeinference instead of declaring SymbolValue?<XXX>-class
• Pager-LR(1)
• Refactor unary productions automatically
• Enhanced Profiling and optimisation of parsing engines
• Attribute grammars and attribute evaluation
• LL(k) or LL(*) generators
• JFlex-API interface

Enjoy!