Install Node 14 or later. ANTLR 4 now requires Node 14+ for ES module support.
Install the package with npm in your project.
npm install antlr4-vensim
Import the lexer, parser, and/or visitor.
import { ModelLexer, ModelParser, ModelVisitor } from 'antlr4-vensim'
Refer to the ANTLR 4 JavaScript runtime documentation for further details about using the parser.
The grammar development tools in this package require macOS. They are not required to use the parser.
Vensim grammar development uses the ANTLR 4 parser generator.
Install Node 14 or later. ANTLR 4 now requires Node 14+ for ES module support.
Install the latest Java SE JDK. We are only using Java for development purposes, which is covered by the license for Java SE.
Install ANTLR 4 Java tools version 4.12.0.
cd /usr/local/lib
sudo curl -O https://www.antlr.org/download/antlr-4.12.0-complete.jar
Set up ANTLR 4 in .bash_profile.
export CLASSPATH=".:/usr/local/lib/antlr-4.12.0-complete.jar:$CLASSPATH"
Build the parser after modifying the grammar.
pnpm i
pnpm build
The build process generates a Vensim parser in JavaScript in the parser directory.
ModelLexer.js
ModelParser.js
ModelVisitor.js
The tests run against the generated parser in the parser directory, so run pnpm build first if you have changed the grammar.
pnpm test
The tests in tests/expr-precedence.test.js pin down operator grouping in the expr rule. Precedence there comes from the order of the alternatives in grammar/Expr.g4, so reordering them changes how expressions evaluate without causing a parse error.
The a4 script generates and runs a lexer and parser in Java. It is useful in development when you are trying to understand how the lexer is tokenizing your model and what parse tree the parser is generating.
./a4 generate
Generate the parser in Java (not JavaScript) in the java directory.
./a4 lex <mdl-file-pathname>
Run the lexer on the .mdl file to print the token stream. The generate command must be run first.
./a4 tree <mdl-file-pathname>
Run the parser on the .mdl file to print the parse tree. The generate command must be run first.
./a4 clean
Clean out the generated parser files in the java directory.
./a4 tokens
Print the lexer token list.