RuleBasedParser is a fast and sophisticated parser that is at the core of Umple.
It reads in grammar files written in a modified/extended EBNF syntax and produces rules that can be used to parse files written in the corresponding language.
Currently, this can only be compiled as part of the main Umple build. This will be changed in the future to be an entirely independently compilable project with a separate ant build target.
The umple source is available in
src/, and -- once compiled -- the java source is available in
How To Use
Using the RuleBasedParser is simple:
Import the package for the RuleBasedParser In Umple this is
depend cruise.umple.parser.analysis.RuleBasedParser. In Java, just
Construct a parser with
RuleBasedParser rbp = new RuleBasedParser()
- Read in any necessary grammar files for the parser with
rbp.addGrammarFile("<filepath>");The grammar file filepath is relative to the root java package.
- [Optional] Assign a handler for linking files. To do this, call
rbp.setLikedFileHandler(...)and pass a class that implements
- [Optional] Assign a handler for generating analyzers. To do this, call
rbp.setAnalyzerGenerator(...)and pass a class that implements
- [Optional] Create parser actions for specific tokens with
rbp.addParserAction(...), passing it the string action name and a class that implements
- Create a
Fileobject of the file to parse (e.g.
File file = new File("somefile");
- Parse the file with
rbp.parse( file ). To retrieve the tokens from the parse, use