Pythonでコンパイラ: PL/0パーサー ref: http://qiita.com/knoguchi/items/ee949989d0a9f04bee6f

ex1.pl0

VAR x, squ;

PROCEDURE square;
BEGIN
   squ := x * x
END;

BEGIN
   x := 1;
   WHILE x <= 10 DO
   BEGIN
      CALL square;
      ! squ;
      x := x + 1;
   END
END.

file10.txt

condition = "odd" expression |
            expression ("="|"#"|"<"|"<="|">"|">=") expression .

file11.py

condition = ODD + expression | expression + oneOf('= # < <= > >=') + expression

file12.txt

>>> condition.parseString('odd 1')
(['ODD', '1'], {})
>>> condition.parseString('3 <= 1')
(['3', '<=', '1'], {})

file13.py

assign_statement = ident + ":=" + expression

file14.py

call_statement = CALL + ident

file15.py

statement = Forward()
if_statement = IF + condition + THEN + statement

file16.py

while_statement = WHILE + condition + DO + statement

file17.txt

statement = [ ident ":=" expression | "call" ident |
            "begin" statement {";" statement } "end" |
            "if" condition "then" statement |
            "while" condition "do" statement ].

file18.py

statement = Optional(assign_statement
                   | call_statement
                   | BEGIN + statement + ZeroOrMore(";" + statement) + END
                   | if_statement
                   | while_statement
                   )

file19.py

const = CONST + ident + "=" + number + ZeroOrMore("," + ident + "=" + number) + ";"

file2.py

CONST = CaselessKeyword('CONST')
VAR = CaselessKeyword('VAR')
 :

file20.py

var = VAR + ident + ZeroOrMore("," + ident) + ";"

file21.py

block = Forward()
procedure = PROCEDURE + ident + ";" + block + ";"

file22.py

block << Optional(const) + Optional(var) + ZeroOrMore(procedure) + statement

file23.txt

program = block + "."

file24.txt

$ python pl0_parser.py ex1.pl0
Traceback (most recent call last):
  File "pl0_parser.py", line 64, in <module>
    print program.parseString(txt)
  File "/usr/lib/python2.7/dist-packages/pyparsing.py", line 1041, in parseString
    raise exc
pyparsing.ParseException: Expected "." (at char 59), (line:8, col:1)

file25.txt

>>> statement.parseString('''\
... BEGIN
...    x := 1;
...    WHILE x <= 10 DO
...    BEGIN
...       CALL square;
...       ! squ;
...       x := x + 1;
...    END
... END
... ''')
([], {})

file26.txt

$ python pl0_parser.py ex1.pl0
['VAR', 'x', ',', 'squ', ';', 'PROCEDURE', 'square', ';', 'BEGIN', 'squ', ':=', 'x', '*', 'x', 'END', ';', 'BEGIN', 'x', ':=', '1', ';', 'WHILE', 'x', '<=', '10', 'DO', 'BEGIN', 'CALL', 'square', ';', 'x', ':=', 'x', '+', '1', ';', 'END', 'END', '.']

file3.py

from pyparsing import CaselessKeyword, MatchFirst
(CONST, VAR, PROCEDURE, CALL, BEGIN, END, IF, THEN, WHILE, DO, ODD) = map(CaselessKeyword,
"CONST, VAR, PROCEDURE, CALL, BEGIN, END, IF, THEN, WHILE, DO, ODD".replace(",","").split())

keyword = MatchFirst((CONST, VAR, PROCEDURE, CALL, BEGIN, END, IF, THEN, WHILE, DO, ODD))

file4.txt

>>> print keyword.parseString('CONST')
['CONST']
>>> print keyword.parseString('const')
['CONST']

file5.txt

from pyparsing import Word, alphas, alphanums
ident = ~keyword + Word(alphas, alphanums+"_")

file6.txt

>>> print repr(ident.parseString('valid_id'))
(['valid_id'], {})
>>> print repr(ident.parseString('0123bad_id'))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/python2.7/dist-packages/pyparsing.py", line 1041, in parseString
    raise exc
pyparsing.ParseException: Expected W:(abcd...,abcd...) (at char 0), (line:1, col:1)
>>> print repr(ident.parseString('CONST'))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/python2.7/dist-packages/pyparsing.py", line 1041, in parseString
    raise exc
pyparsing.ParseException: Found unwanted token, {"CONST" | "VAR" | "PROCEDURE" | "CALL" | "BEGIN" | "END" | "IF" | "THEN" | "WHILE" | "DO" | "ODD"} (at char 0), (line:1, col:1)

file7.txt

expression = [ "+"|"-"] term { ("+"|"-") term}.
term = factor {("*"|"/") factor}.
factor = ident | number | "(" expression ")"

file8.py

number = Regex(r"\d+(\.\d*)?([eE][+-]?\d+)?")
term = Forward()
factor = Forward()
expression = Optional(oneOf("+ -")) + term + ZeroOrMore( oneOf("+ -") + term)
term << factor + ZeroOrMore(oneOf("* /") + factor)
factor << ident | number | "(" + expression + ")"

file9.txt

>>> expression.parseString('123')
(['123'], {})
>>> expression.parseString('123+456')
(['123', '+', '456'], {})
>>> expression.parseString('(x+y)*z')
(['(', 'x', '+', 'y', ')', '*', 'z'], {})

pl0_bnf.txt

ident = alpha { alpha | number | '_' } .

program = block "." .

block = [ "const" ident "=" number {"," ident "=" number} ";"]
        [ "var" ident {"," ident} ";"]
        { "procedure" ident ";" block ";" } statement .
statement = [ ident ":=" expression | "call" ident |
            "begin" statement {";" statement } "end" |
            "if" condition "then" statement |
            "while" condition "do" statement ].
condition = "odd" expression |
            expression ("="|"#"|"<"|"<="|">"|">=") expression .
expression = [ "+"|"-"] term { ("+"|"-") term}.
term = factor {("*"|"/") factor}.
factor = ident | number | "(" expression ")".

pl0_parser.py

from pyparsing import CaselessKeyword, MatchFirst, Word, alphas, alphanums, Forward, Optional, oneOf, ZeroOrMore, Regex


# 1. reserved keyword
(CONST, VAR, PROCEDURE, CALL, BEGIN, END, IF, THEN, WHILE, DO, ODD) = map(CaselessKeyword,
"CONST, VAR, PROCEDURE, CALL, BEGIN, END, IF, THEN, WHILE, DO, ODD".replace(",", "").split())
keyword = MatchFirst((CONST, VAR, PROCEDURE, CALL, BEGIN, END, IF, THEN, WHILE, DO, ODD))

# 2. identifier
ident = ~keyword + Word(alphas, alphanums + "_")

# 3. expression
number = Regex(r"\d+(\.\d*)?([eE][+-]?\d+)?")
term = Forward()
factor = Forward()
expression = Optional(oneOf("+ -")) + term + ZeroOrMore(oneOf("+ -") + term)
term << (factor + ZeroOrMore(oneOf("* /") + factor))
factor << (ident | number | "(" + expression + ")")

# 4. condition
condition = ODD + expression | expression + oneOf('= # < <= > >=') + expression

# 5. assignment
assign_statement = ident + ":=" + expression

# 6. call
call_statement = CALL + ident

# 7. if-then
statement = Forward()
if_statement = IF + condition + THEN + statement

# 8. while-do
while_statement = WHILE + condition + DO + statement

# 9. statement
statement << Optional(assign_statement
                      | call_statement
                      | BEGIN + statement + ZeroOrMore(";" + statement) + END
                      | if_statement
                      | while_statement
)

# 10. const
const = CONST + ident + "=" + number + ZeroOrMore("," + ident + "=" + number) + ";"

# 11. var
var = VAR + ident + ZeroOrMore("," + ident) + ";"

# 12. procedure
block = Forward()
procedure = PROCEDURE + ident + ";" + block + ";"

# 13. block
block << Optional(const) + Optional(var) + ZeroOrMore(procedure) + statement

# 14. program
program = block + "."

if __name__ == '__main__':
    import sys
    with open(sys.argv[1], 'r') as fp:
        txt = fp.read()
        print program.parseString(txt)