g = Grammar(r"""
grammar = _ defn*.
defn : var '='_ exp ';'_ :hug.
exp : 'Choice' args :mk_choice
| 'Fixed' args :mk_fixed
| 'Sequence' args :mk_sequence
| 'Shuffle' args :mk_shuffle
| 'Weighted' args :mk_weighted
| /Period\b/_ :'.'
| /Comma\b/_ :','
| /Semicolon\b/_ :';'
| /Dash\b/_ :'--'
| /AAn\b/_ :'-a-an-'
| /Concat\b/_ :'-adjoining-'
| /null\b/_ :'()'
| var
| string
| int.
args : '('_ exps? ')'_.
exps : exp (','_ exps)*.
var : /([A-Za-z_]\w*)/_ :mk_var.
int : /(\d+)/ :int.
string : '"' qchar* '"'_ :join.
qchar : !/["\\]/ /(.)/.
_ : (space | comment)*.
space : /\s+/.
comment : '/*' (!'*/' anyone)* '*/'.
anyone : /./ | /\n/. # Ugh.
""")(
mk_var=lambda s: '-' + '-'.join(parse_camel(s)) + '-',
mk_choice=lambda *xs: ' / '.join(xs),
mk_fixed=lambda tag, choice: '%s{ %s }' % (tag, choice),
mk_sequence=lambda *xs: ' '.join(map(wrap, xs)),
mk_shuffle=lambda *xs: '{ %s }' % (' / '.join(xs)),
mk_weighted=lambda *spairs: ' / '.join(
'[%s] %s' % (w, wrap(x)) for w, x in zip(spairs[0::2], spairs[1::2])),
)
STRING ~: /"([^"]*)"/ FNORD # XXX
| /'([^']*)'/ FNORD.
IDENT ~: !RESERVED_WORD {IdentifierName} FNORD.
# XXX incomplete
IdentifierName ~= IdentifierStart IdentifierPart*.
IdentifierStart ~= UnicodeLetter
| '$'
| '_'.
IdentifierPart ~= IdentifierStart.
UnicodeLetter ~= /[A-Za-z]/.
"""
#import sys; sys.setrecursionlimit(5000)
gr = Grammar(g)
import microses
grr = gr.bind(microses).expecting_one_result()
def test(filename):
from pprint import pprint
print 'testing', filename
with open(filename) as f:
text = f.read()
result = grr(text)
for form in result:
pprint(form.as_sexpr())
## grr('a.b =c;')
#. (Assign(Get(Variable('a'), 'b'), '=', Variable('c')),)
# Following http://www.json.org/
json_parse = Grammar(r"""
start : _ value.
object : '{'_ pairs? '}'_ :mk_object.
pairs : pair (','_ pair)*.
pair : string ':'_ value :hug.
array : '['_ elements? ']'_ :hug.
elements : value (','_ value)*.
value : string | number
| object | array
| /(true|false|null)\b/_ :mk_literal.
string : '"' char* '"'_ :join.
char : /([^\x00-\x1f"\\])/
| /\\(["\/\\])/
| /(\\[bfnrt])/ :escape
| /(\\u)/ xd xd xd xd :join :escape.
xd : /([0-9a-fA-F])/.
number : int (frac exp? | exp)? _ :join :mk_number.
int : /(-?0)/ !/\d/
| /(-?[1-9]\d*)/.
frac : /([.]\d+)/.
exp : /([eE][+-]?\d+)/.
_ : /\s*/.
""")(**globals()).start
toy_grammar = Grammar(r"""
main : _ E !/./.
E : Fp '`'_ V '`'_ E :fold_infix_app
| Fp :fold_apps
| '&'_ Vp '=>'_ E :fold_lam
| /let\b/_ Decls E :make_let
| /case\b/_ E Cases :make_case.
Cases : Case+ :hug.
Case : '|'_ Param '=>'_ E :hug.
Param : Const
| V
| '('_ Param ')'_
| '['_ ParamList ']'_.
ParamList : Param ','_ Param :make_list_pattern.
Decls : Decl+ :hug.
Decl : /defer\b/_ V ';'_ :make_defer
| /bind\b/_ V '='_ E ';'_ :make_bind
| Vp '='_ E ';'_ :make_eqn.
Fp : F+ :hug.
F : Const :make_const
| V :make_var
| '('_ E ')'_
| '{'_ F Fp '}'_ :fold_send
| '['_ EList ']'_ :hug :make_list_expr.
EList : (E (','_ EList)?)?.
Vp : V+ :hug.
V : Identifier
| Operator.
Identifier : /(?!let\b|case\b|defer\b|bind\b)([A-Za-z_]\w*)\b\s*/.
Operator : /(<=|:=|[!+-.])\s*/.
Const : '.'_ V :make_lit_sym
| /"([^"]*)"/_ :repr
| /(-?\d+)/_
| '('_ ')'_ :'()'
| '['_ ']'_ :'[]'.
_ : /\s*/.
""")(**globals())
"""
The customary calculator example.
"""
import operator
from parson import Grammar
g = Grammar(r"""
top : _ exp0 !/./.
exp0 : exp1 ( '+'_ exp1 :add
| '-'_ exp1 :sub)*.
exp1 : exp2 ( '*'_ exp2 :mul
| '//'_ exp2 :div
| '/'_ exp2 :truediv
| '%'_ exp2 :mod)*.
exp2 : exp3 ( '^'_ exp2 :pow)?.
exp3 : '('_ exp0 ')'_
| '-'_ exp1 :neg
| /(\d+)/_ :int.
_ = /\s*/.
""")(**operator.__dict__)
## g.top('42 * (5-3) + -2^2')
#. (80,)
## g.top('2^3^2')
#. (512,)
## g.top('5-3-1')
#. (1,)
| 'b' A
| .
A : 'a' S
| 'b' A A.
B : 'b' S
| 'a' B B.
""")()
## as_and_bs.allS("abaabbbbaa")
#. ()
nums = Grammar(r"""
allnums : nums? !/./.
nums : num (',' num)*.
num : /(\d+)/ :int.
""")()
sum_nums = lambda s: sum(nums.allnums(s))
## sum_nums('10,30,43')
#. 83
one_word = Grammar(r"word : /\w+/ :position.")()
## one_word.word('hello')
#. (5,)
## one_word.word('hello there')
#. (5,)
## one_word.word.attempt(' ')
grammar = Grammar(r""" box* :end.
box: name '{' [stmt* :hug] '}' :Box.
stmt: "var" name ++ ',' ';' :hug :Decl
| "put" [name ':' | :None] box ';'? :Put
| "conn" expr ("to" expr)+ :hug
("using" expr box '<' expr ',' expr '>' ';' :Pen
| ';' :Conn)
| "spline" expr ("to" expr)+ ';' :hug :Spline
| "compass" expr "to" expr ';' :Compass
| justify string "at" expr ';' :Text
| expr ('=' expr)+ ';' :hug :Equate
| expr ('~' expr)+ ';' :hug :Default.
justify ~: {"left"|"right"|"center"} FNORD
| :'center'.
expr: term ( '+' term :Add
| '-' term :Sub )*.
term: factor ( '*' factor :Mul
| '/' factor :Div )*.
factor: atom ('[' expr ',' expr ']' :Relatively)*.
atom: '(' number ',' number ')' :complex :Literal
| number :complex :Literal
| '-' atom :Negate
| '(' expr ')'
| unaryfn '(' expr ')' :CallPrim
| name :Ref ('.' name :Of)*.
unaryfn: "abs" :Abs
| "cis" :Cis
| "unit" :Unit.
name: /([A-Za-z_][A-Za-z_0-9]*)/.
number ~: { '-'? (/\d*/ '.' /\d+/ | /\d+/) } FNORD :float.
string ~: '"' {/[^\\"]*/} '"' FNORD.
FNORD ~: (/\s+/ | comment)*.
comment ~: '/*' (!'*/' :anyone)* '*/'.
""")
grammar = Grammar(r"""
command : /(\d+)/_ :int /(.*)/ /$/ :set_line
| 'run' /\b/_ /$/ :run
| 'new' /\b/_ /$/ :new
| 'load' /\b/_ /(\S+)/_ /$/ :load
| 'save' /\b/_ /(\S+)/_ /$/ :save
| stmt
| /$/.
stmt : 'print' /\b/_ printing /$/ :next
| '?' _ printing /$/ :next
| 'input' /\b/_ id /$/ :input :next
| 'goto' /\b/_ exp0 /$/ :goto
| 'if' /\b/_ relexp 'then' /\b/_ exp0 /$/ :if_goto
| 'gosub' /\b/_ exp0 /$/ :gosub
| 'return' /\b/_ /$/ :return_
| 'end' /\b/_ /$/ :end
| 'list' /\b/_ /$/ :list :next
| 'rem' /\b/ /.*/ /$/ :next
| ('let' /\b/_)? id '='_ exp0 /$/ :store :next.
printing : (display writes)?.
writes : ';'_ printing
| ','_ :space printing
| :newline.
display : exp0 :write
| '"' [qchar :write]* '"'_.
qchar : /"(")/
| /([^"])/.
relexp : exp0 ( '<>'_ exp0 :ne
| '<='_ exp0 :le
| '<' _ exp0 :lt
| '=' _ exp0 :eq
| '>='_ exp0 :ge
| '>' _ exp0 :gt
)?.
exp0 : exp1 ( '+'_ exp1 :add
| '-'_ exp1 :sub
)*.
exp1 : exp2 ( '*'_ exp2 :mul
| '/'_ exp2 :idiv
)*.
exp2 : primary ('^'_ exp2 :pow)?.
primary : '-'_ exp1 :neg
| /(\d+)/_ :int
| id :fetch
| '('_ exp0 ')'_.
id : /([a-z])/_. # TODO: longer names, screening out reserved words
_ : /\s*/.
""")
exp3 = delay(lambda: g.exp3)
exp1 = PrececedenceParser(exp3, [
LeftAssoc(('*'+_, mul), ('//'+_, div), ('/'+_, truediv), ('%'+_, mod)),
RightAssoc(('^'+_, pow)),
])
exps = PrececedenceParser(exp1, [
LeftAssoc(('+'+_, add), ('-'+_, sub)),
])
g = Grammar(r"""
top = _ :exps !/./.
exp3 : '('_ :exps ')'_
| '-'_ :exp1 :neg
| /(\d+)/_ :int.
_ = /\s*/.
""")(**globals())
## g.top('42 *(5-3) + -2^2')
#. (80,)
## g.top('2^3^2')
#. (512,)
## g.top('5-3-1')
#. (1,)
## g.top('3//2')
#. (1,)
## g.top('3/2')
#. (1.5,)
def p(grammar, rule, text):
parse = getattr(Grammar(grammar)(**globals()), rule)
try:
return parse(text)
except Unparsable, e:
return e
Ns_x = set(len(m) for m in xmatches)
Ns_y = set(n - m for n in Ns for m in Ns_x if n - m >= 0)
ymatches = y(Ns_y)
return set(m1 + m2 for m1 in xmatches for m2 in ymatches
if len(m1 + m2) in Ns)
grammar = Grammar(r"""
regex : exp !/./.
exp : term ('|' exp :either)*
| :empty.
term : factor (term :chain)*.
factor : primary ( '*' :star
| '+' :plus
| '?' :optional
)?.
primary : '(' exp ')'
| '[' char* ']' :join :oneof
| '.' :dot
| /\\(.)/ :literal
| /([^.()*+?|[\]])/ :literal.
char : /\\(.)/
| /([^\]])/.
""")
parser = grammar(**globals())
## generate('.+', range(5))
#. ['?', '??', '???', '????']
## generate('a[xy]+z()*|c.hi', range(5))
#. ['axz', 'ayz', 'axxz', 'axyz', 'ayxz', 'ayyz', 'c?hi']
## generate('(Chloe|Yvette), a( precocious)? (toddler|writer)', range(28))
"""
A bunch of small examples, some of them from the LPEG documentation.
Crudely converted from Peglet. TODO: make them nicer.
"""
from parson import Grammar, Unparsable, exceptionally
parse_words = Grammar(r'words : /\W*(\w+)/ words | .')()
# The equivalent re.split() would return extra '' results first and last:
## parse_words.words('"Hi, there", he said.')
#. ('Hi', 'there', 'he', 'said')
def Tag(label):
return lambda *parts: (label, ) + parts
name = Grammar(r"""
name : title first middle last.
title : (/(Dr|Mr|Ms|Mrs|St)[.]?/ | /(Pope(?:ss)?)/) _ :Title |.
first : /([A-Za-z]+)/ _ :First.
middle : (/([A-Z])[.]/ | /([A-Za-z]+)/) _ :Middle |.
last : /([A-Za-z]+)/ :Last.
_ : /\s+/.
""")(Title=Tag('title'),
First=Tag('first'),
Middle=Tag('middle'),
Last=Tag('last'))
## name.name('Popess Darius Q. Bacon')
#. (('title', 'Popess'), ('first', 'Darius'), ('middle', 'Q'), ('last', 'Bacon'))
"""
Convert a Peglet grammar to a Parson one.
"""
import re
from parson import Grammar, alter
name = r'[A-Za-z_]\w*'
grammar = Grammar(r"""
grammar : _? rule* !/./.
rule : name _ '= ' :equ token* :'.' _?.
token : '|' :'|'
| /(\/\w*\/\s)/
| name !(_ '= ')
| '!' :'!'
| _ !(name _ '= ' | !/./)
| !('= '|name) /(\S+)/ :mk_regex.
name : /(""" + name + """)/ !!(/\s/ | !/./).
_ : /(\s+)/.
""")
def mk_regex(s):
return '/' + s.replace('/', '\\/') + '/'
def peglet_to_parson(text):
nonterminals = set()
def equ(name, space):
ProcedureDeclaration: ProcedureHeading ';'_ ProcedureBody ident.
declarations: ('CONST'_ (ident '='_ expression ';'_)*)?
('TYPE'_ (ident '='_ type ';'_)*)?
('VAR'_ (IdentList ':'_ type ';'_)*)?
(ProcedureDeclaration ';'_)*.
module: 'MODULE'_ ident ';'_ declarations
('BEGIN'_ StatementSequence)? 'END'_ ident '.'_.
_: whitespace*.
whitespace: /\s+/ | comment.
comment: '(*' commentchunk* '*)'.
commentchunk: comment | !'*)' /.|\n/. # XXX are comments nested in Oberon-0?
keyword: /BEGIN|END|MODULE|VAR|TYPE|CONST|PROCEDURE|RECORD|ARRAY|OF|WHILE|DO|IF|ELSIF|THEN|ELSE|OR|DIV|MOD/ /\b/.
top: _ module !/./.
"""
grammar = Grammar(grammar_source)()
# TODO test for expected parse failures
## from parson import exceptionally
## import glob
## for filename in sorted(glob.glob('ob-bad/*.ob')): print exceptionally(lambda: test(filename))
#. testing ob-bad/badassign.ob
#. (top, 'MODULE badassign;\n\nBEGIN\n ', '1 := 2\nEND badassign.\n')
#. testing ob-bad/commentnoend.ob
#. (top, "MODULE commentnoend;\n (* started off well,\n but didn't finish\nEND commentnoend.\n", '')
#. testing ob-bad/keywordasname.ob
#. (top, 'MODULE ', 'END;\nEND END.\n')
#. testing ob-bad/repeatsection.ob
#. (top, 'MODULE repeatsection;\n\nCONST\n aconst = 10;\n\n', 'CONST\n aconst = 20;\n\nEND repeatsection.\n')
qstring: /'([^']*)'/_.
tokenid: /([A-Z_]+)/_.
id: /([a-z_]+)/_.
_ = /\s*/.
"""
def foldr1(f, xs):
return xs[0] if len(xs) == 1 else f(xs[0], foldr1(f, xs[1:]))
def Chain(*pes): return empty if not pes else foldr1(chain, pes)
def Optional(pe): return pe.maybe()
def Literal(s): return label(one_that(lambda t: t[1] == s), repr(s)) # XXX
def Token(name): return label(one_that(lambda t: t[1] == name), name) # XXX
def RuleRef(name): return delay((lambda: rules[name]), name)
grammar = Grammar(g)(**globals())
subset = open('subset').read()
metagrammar = grammar.grammar(subset)
rules = dict(metagrammar)
pygrammar = rules['file_input']
## pygrammar([('', 'ENDMARKER',)])
#. ()
## grammar.grammar("dotted_name: NAME ('.' NAME)*")
#. (('dotted_name', (NAME (('.' NAME))*)),)
## for pair in grammar.grammar('yo: hey boo: yah'): print pair
#. ('yo', hey)
#. ('boo', yah)
grammar = Grammar(r"""
program: _ (definition | declaration)* ('__END__' | :end).
definition: defheader '{'_ [declaration* :hug] '}'_ :Definition.
defheader: 'define'__ ID ['extends'__ ID | :None].
declaration: ID declarators ';'_ :VarDecl
| constraint_section
| draw_section.
declarators: declarator (','_ declarator)* :hug.
declarator: ID [('('_ params ')'_)? :hug] :Declarator.
params = param_spec (','_ param_spec)*.
param_spec: ID '='_ expression :hug.
constraint_section: 'constraints'__ '{'_ constraint* '}'_ :hug :Constraints.
constraint: expression '='_ expression ';'_ :hug.
draw_section: 'draw'__ '{'_ drawable* '}'_ :hug :Draw.
drawable: name ';'_ :DrawName
| '&'_ ID ';'_ :DrawFunction.
expression: term ('+'_ term :Add
|'-'_ term :Sub)*.
term: atom ('*'_ atom :Mul
|'/'_ atom :Div)*.
atom: name
| tuple
| NUMBER :Number
| '-'_ expression :Negate
| '('_ expression ')'_.
name: ID :Name ('.'_ ID :Dot)*.
tuple: '('_ expression (','_ expression)+ ')'_ :hug :Tuple.
# Lexical grammar
NUMBER: { mantissa /[eE]\d+/? } _ :float.
mantissa = /\d+/ ('.' /\d*/)?
| '.' /\d+/.
ID = /([a-zA-Z_]\w*)/ _. # XXX need to rule out keywords?
__ = /\b/_. # (i.e. a keyword must match up to a word boundary)
_ = /\s*/.
""")