pyLRp.pyLRp

# pyLRp grammar file for pyLRp
# this must be kept parsable with the limited bootstrap parser
# it is the definite reference on the format

%lexer


\#.*\n+ NEWLINE

\n\s+ NEWLINE
# no lexer stater are available in the pre-parser,
# space matters here and there (python code ...)
# therefore anyone who is whitespace ignorant: please notiy us
\s+ SPACE

\%restart RESTART
\%lexer LEXER
\%parser PARSER

.* ARBLINE

\".*\" TEXT
[A-Za-z_][A-Za-z0-9_]* SYMBOL

\<[A-Z]+\> STATE
# this could be more concrete
([^ \t]|\\ )+ REGEX

%parser

file:
    %empty
    file part

part:
    header
    lexer
    parser

header:
    %empty
    header pycodeline

pycodeline:
    NEWLINE
    SPACE ARBLINE NEWLINE

lexer:
    LEXER SPACE NEWLINE
    lexerline

lexerline:
    STATE SPACE REGEX SPACE lexeraction SPACE NEWLINE
    REGEX SPACE lexeraction SPACE NEWLINE

lexeraction:
    lexeraction1
    lexeraction SPACE "," SPACE lexeraction1

lexeraction1:
    RESTART
    SYMBOL
    BEGIN
    END
    APPEND
    COLLECT

parser:
    PARSER SPACE NEWLINE
    rule

rule:
    SYMBOL SPACE ":" SPACE rulebody

rulebody:
    NEWLINE
    NEWLINE SPACE EMPTY SPACE "{" action "}" SPACE NEWLINE
    NEWLINE SPACE prod  SPACE "{" action "}" SPACE NEWLINE

prod:
    prod1
    prod SPACE prod1

prod1:
    SYMBOL
    TEXT

# clean
action:
    SPACE NEWLINE
    action SPACE ARBLINE NEWLINE