The parser is generated with ANTLR [44] from the grammars of both SPOT and Fortran expressed in EBNF. We chose ANTLR because it is a powerful generator that can not only be used to generate a recognizer for the language, but can also be used to build an AST for the recognized program, which can then be traversed and manipulated.
ALGORITHM 4: Primary EBNF grammar of SPOT programFile :'Transformer' ID '{' transformBody (';' transformBody) * '}' -> ^ (TRANSFORMER_ND ID transformBody+); transformBody :transformScope '{' transformStatement+ '}' -> ^(TFBODY_ND transformScope transformStatement+) |'IncludeCode' '{' statement+ '}' ('into' filename) -> ^ (SOURCE_CODE statement+); transformScope : 'Within' '(' scopeIndicator ID ')' -> ^ ('Within' scopeIndicator ID); scopeIndicator :'Function' |'Module' |'Project' |'Statement' pointIndicator :'FunctionCall' |'VariableRead' |'VariableWrite' | statementTypeName //collect all statements of a type | '"' statement '"';//collect all statements with original source code, e.g.
We provide an EBNF syntax for each extension of SQL-92.
We start with an EBNF syntax for statement modifiers, and then discuss their meanings.
We describe its architecture and components as well as an implemented lesson on the topic "EBNF" (Enhanced Backus-Naur Formalism, a grammar-like technique used for describing the syntax of programming languages).
A summary of all actions necessary for building a SmexWeb application is given in another section where the EBNF example application is presented.
To describe syntax, we use an
EBNF grammar with standard metasymbols for