def make_enewick_parser():
# atoms
lparen = Literal("(").suppress()
rparen = Literal(")").suppress()
colon = Literal(":").suppress()
# semicolon = Literal(";").suppress()
comma = Literal(",").suppress()
point = Literal(".")
e = CaselessLiteral("E")
sharp = Literal("#").suppress()
# terminal
name = Word(
alphanums + alphas8bit + "_" + "-" + "." + "+" + "&" + "/" + "~" + "{" + "}" + "*" + "'" + '"' + "\\" + "?"
)
string = Word(alphas)
fnumber = Combine(
Word("+-" + nums, nums) + Optional(point + Optional(Word(nums))) + Optional(e + Word("+-" + nums, nums))
).setParseAction(cvtReal)
number = Combine(Word(nums)).setParseAction(cvtInt)
label = (
Optional(name).setResultsName("label")
+ Optional(sharp + Optional(string).setResultsName("type") + number.setResultsName("tag"))
+ Optional(colon + fnumber).setResultsName("length")
)
subtree = Forward()
subtreelist = Forward()
subtree << Group(((lparen + subtreelist + rparen).setResultsName("subtree") | label) + Optional(label))
subtreelist << subtree + Optional(comma + subtreelist)
tree = subtree + Word(";").suppress()
return tree.parseString
def define_number(self):
"""
Return the syntax definition for a number in Arabic Numerals.
Override this method to support numeral systems other than Arabic
Numerals (0-9).
Do not override this method just to change the character used to
separate thousands and decimals: Use :attr:`T_THOUSANDS_SEPARATOR`
and :attr:`T_DECIMAL_SEPARATOR`, respectively.
"""
# Defining the basic tokens:
to_dot = lambda t: "."
to_plus = lambda t: "+"
to_minus = lambda t: "-"
positive_sign = Literal(self._grammar.get_token("positive_sign"))
positive_sign.setParseAction(to_plus)
negative_sign = Literal(self._grammar.get_token("negative_sign"))
negative_sign.setParseAction(to_minus)
decimal_sep = Literal(self._grammar.get_token("decimal_separator"))
decimal_sep.setParseAction(to_dot)
thousands_sep = Suppress(self._grammar.get_token("thousands_separator"))
digits = Word(nums)
# Building the integers and decimals:
sign = positive_sign | negative_sign
thousands = Word(nums, max=3) + \
OneOrMore(thousands_sep + Word(nums, exact=3))
integers = thousands | digits
decimals = decimal_sep + digits
number = Combine(Optional(sign) + integers + Optional(decimals))
number.setParseAction(self.make_number)
number.setName("number")
return number
def define_identifier(self):
"""
Return the syntax definition for an identifier.
"""
# --- Defining the individual identifiers:
# Getting all the Unicode numbers in a single string:
unicode_numbers = "".join([unichr(n) for n in xrange(0x10000)
if unichr(n).isdigit()])
unicode_number_expr = Regex("[%s]" % unicode_numbers, re.UNICODE)
space_char = re.escape(self._grammar.get_token("identifier_spacing"))
identifier0 = Regex("[\w%s]+" % space_char, re.UNICODE)
# Identifiers cannot start with a number:
identifier0 = Combine(~unicode_number_expr + identifier0)
identifier0.setName("individual_identifier")
# --- Defining the namespaces:
namespace_sep = Suppress(self._grammar.get_token("namespace_separator"))
namespace = Group(ZeroOrMore(identifier0 + namespace_sep))
namespace.setName("namespace")
# --- The full identifier, which could have a namespace:
identifier = Combine(namespace.setResultsName("namespace_parts") +
identifier0.setResultsName("identifier"))
identifier.setName("full_identifier")
return identifier
def func_tokens(dictionary, parse_action):
func_name = Word(alphas+'_', alphanums+'_')
func_ident = Combine('$' + func_name.copy()('funcname'))
func_tok = func_ident + originalTextFor(nestedExpr())('args')
func_tok.leaveWhitespace()
func_tok.setParseAction(parse_action)
func_tok.enablePackrat()
rx_tok = Combine(Literal('$').suppress() + Word(nums)('num'))
def replace_token(tokens):
index = int(tokens.num)
return dictionary.get(index, u'')
rx_tok.setParseAction(replace_token)
strip = lambda s, l, tok: tok[0].strip()
text_tok = CharsNotIn(u',').setParseAction(strip)
quote_tok = QuotedString('"')
if dictionary:
arglist = Optional(delimitedList(quote_tok | rx_tok | text_tok))
else:
arglist = Optional(delimitedList(quote_tok | text_tok))
return func_tok, arglist, rx_tok
def expression(self):
from pyparsing import Suppress,Combine,Optional,oneOf,OneOrMore,Word,nums,Group,alphas,alphanums,Literal,SkipTo,empty,lineEnd
cvtInt = lambda toks: int(toks[0])
cvtReal = lambda toks: float(toks[0])
cvtTuple = lambda toks : tuple(toks.asList())
nameJoin = lambda toks : "".join([tok.replace("#","") for tok in toks[0]])
#lambda toks: " ".join([str(t) for t in toks[0]])
# define punctuation as suppressed literals
lparen,rparen,lbrack,rbrack,lbrace,rbrace,colon = map(Suppress,"()[]{}:")
integer = Combine(Optional(oneOf("+ -")) + Word(nums))\
.setName("integer")\
.setParseAction( cvtInt )
real = Combine(Optional(oneOf("+ -")) + Word(nums) + "." +
Optional(Word(nums)) +
Optional(oneOf("e E")+Optional(oneOf("+ -")) +Word(nums))).setName("real").setParseAction( cvtReal )
# TREE DEFINITION
# ((seq2: 0.537243, seq1: 0.000004): 0.255741, seq3: 0.281503);
tree_w_branches = (
OneOrMore(Word("():,."+alphas+nums))+Literal(";")
).setParseAction(lambda tokens: " ".join(tokens[:-1])+";")
# SITE PROBABILITIES
# site Freq Data:
# 1 1 AAA: A(0.978) A(1.000)
site_prob = (
integer.setResultsName("site",listAllMatches=True) +
integer.setResultsName("freq",listAllMatches=True) +
Word(alphas+"-").setResultsName("extant",listAllMatches=True) + colon +
Group(OneOrMore(Group(Word(alphas,exact=1)+lparen+real+rparen))).setResultsName("probability",listAllMatches=True) +
lineEnd
)
# ANCESTRAL SEQUENCES
# seq1 ACC
# node #4 ACC
# Optional # character with node # needs to be joined into a single name
sequence = (
Group(Word(alphanums)+
Optional(Combine(Literal("#")+Word(nums)))).setParseAction(nameJoin).setResultsName("name",listAllMatches=True)+
Word(alphas+"- ").setResultsName("sequence", listAllMatches=True)+lineEnd
)
return (SkipTo(Literal("Ancestral reconstruction by AAML."),include=True).suppress() +
tree_w_branches.setResultsName("tree") +
SkipTo(Literal("site")+Literal("Freq")+Literal("Data:"), include=True,).suppress()+
Group(OneOrMore(site_prob)).setResultsName("sites")+
SkipTo(Literal("List of extant and reconstructed sequences")+Word(nums)+Word(nums), include=True).suppress()+
Group(OneOrMore(sequence)).setResultsName("sequences")+
SkipTo(Literal("for a site."),include=True).suppress()+
Group(OneOrMore(real)).setResultsName("probability")+
empty
)
def is_ipv4_addr(inputstr):
from pyparsing import Combine, Word, nums
ipAddress = Combine(Word(nums) + ('.' + Word(nums)) * 3)
try:
ipAddress.parseString(inputstr)
return True
except:
return False
def _BNF(self):
base16 = Literal("$")
hex = Combine(base16 + Word(hexnums + "_"))
base4 = Literal("%%")
quaternary = Combine(base4 + Word("0123_"))
base2 = Literal("%")
binary = Combine(base2 + Word("01_"))
plusminus = Literal("+") | Literal("-")
integer = Combine(Optional(plusminus) + Word(nums+"_"))
name_token = Combine(Optional(Literal(":") | Literal("@")) + Word("_" + alphas, "_" + alphanums))
name_token.setParseAction(self._mark_name_token)
lparens = Literal("(").suppress()
rparens = Literal(")").suppress()
# op0 = Literal("@")
op1 = (Literal("^^") | Literal("||") | Literal("|<") | Literal(">|") | Literal("!")).setParseAction(self._mark_unary)
op2 = Literal("->") | Literal("<-") | Literal(">>") | Literal("<<") | Literal("~>") | Literal("><")
op3 = Literal("&")
op4 = Literal("|") | Literal("^")
op5 = Literal("**") | Literal("*") | Literal("//") | Literal("/")
op6 = Literal("+") | Literal("-")
op7 = Literal("#>") | Literal("<#")
op8 = Literal("<") | Literal(">") | Literal("<>") | Literal("==") | Literal("=<") | Literal("=>")
op9 = Literal("NOT").setParseAction(self._mark_unary)
op10 = Literal("AND")
op11 = Literal("OR")
op12 = Literal(",")
expr = Forward()
atom = name_token | hex | quaternary | binary | integer | quotedString
atom.setParseAction(self._push)
atom = atom | (lparens + expr.suppress() + rparens)
# term0 = atom + ZeroOrMore((op0 + atom) .setParseAction(self._push))
# term1 = term0 + ZeroOrMore((op1 + term0) .setParseAction(self._push))
term1 = atom + ZeroOrMore((op1 + atom) .setParseAction(self._push))
term2 = term1 + ZeroOrMore((op2 + term1) .setParseAction(self._push))
term3 = term2 + ZeroOrMore((op3 + term2) .setParseAction(self._push))
term4 = term3 + ZeroOrMore((op4 + term3) .setParseAction(self._push))
term5 = term4 + ZeroOrMore((op5 + term4) .setParseAction(self._push))
term6 = term5 + ZeroOrMore((op6 + term5) .setParseAction(self._push))
term7 = term6 + ZeroOrMore((op7 + term6) .setParseAction(self._push))
term8 = term7 + ZeroOrMore((op8 + term7) .setParseAction(self._push))
term9 = term8 + ZeroOrMore((op9 + term8) .setParseAction(self._push))
term10 = term9 + ZeroOrMore((op10 + term9) .setParseAction(self._push))
term11 = term10 + ZeroOrMore((op11 + term10).setParseAction(self._push))
expr << term11 + ZeroOrMore((op12 + term11).setParseAction(self._push))
return expr
def getkw_bnf(self):
sect_begin = Literal("{").suppress()
sect_end = Literal("}").suppress()
array_begin = Literal("[").suppress()
array_end = Literal("]").suppress()
tag_begin = Literal("<").suppress()
tag_end = Literal(">").suppress()
eql = Literal("=").suppress()
dmark = Literal('$').suppress()
end_data=Literal('$end').suppress()
prtable = alphanums+r'!$%&*+-./<>?@^_|~'
ival=Regex('[-]?\d+')
dval=Regex('-?\d+\.\d*([eE]?[+-]?\d+)?')
lval=Regex('([Yy]es|[Nn]o|[Tt]rue|[Ff]alse|[Oo]n|[Oo]ff)')
# Helper definitions
kstr= quotedString.setParseAction(removeQuotes) ^ \
dval ^ ival ^ lval ^ Word(prtable)
name = Word(alphas+"_",alphanums+"_")
vec=array_begin+delimitedList(dval ^ ival ^ lval ^ Word(prtable) ^ \
Literal("\n").suppress() ^ \
quotedString.setParseAction(removeQuotes))+array_end
sect=name+sect_begin
tag_sect=name+Group(tag_begin+name+tag_end)+sect_begin
# Grammar
keyword = name + eql + kstr
vector = name + eql + vec
data=Combine(dmark+name)+SkipTo(end_data)+end_data
section=Forward()
sect_def=(sect | tag_sect ) #| vec_sect)
input=section | data | vector | keyword
section << sect_def+ZeroOrMore(input) + sect_end
# Parsing actions
ival.setParseAction(self.conv_ival)
dval.setParseAction(self.conv_dval)
lval.setParseAction(self.conv_lval)
keyword.setParseAction(self.store_key)
vector.setParseAction(self.store_vector)
data.setParseAction(self.store_data)
sect.setParseAction(self.add_sect)
tag_sect.setParseAction(self.add_sect)
sect_end.setParseAction(self.pop_sect)
bnf=ZeroOrMore(input) + StringEnd().setFailAction(parse_error)
bnf.ignore(pythonStyleComment)
return bnf
def receiver_input_rule():
path = Combine(ZeroOrMore(word + ".") + word)
input = path.setResultsName("input")
operator = oneOf(operators.keys()).setResultsName("operator")
value = path.setResultsName("value")
comparison = operator + value
is_or_was = Word("is") | Word("was")
condition = Group(input + is_or_was.setResultsName("temporal") + comparison)
res = ZeroOrMore(condition + _and) + condition
conditions = Group(res).setResultsName("conditions")
return Optional("always").setResultsName("always_fire_rule") + when + conditions + then + actions
def parser():
global _parser
if _parser is None:
ParserElement.setDefaultWhitespaceChars("")
lbrack = Literal("[")
rbrack = Literal("]")
lbrace = Literal("{")
rbrace = Literal("}")
lparen = Literal("(")
rparen = Literal(")")
reMacro = Suppress("\\") + oneOf(list("dwsZ"))
escapedChar = ~reMacro + Combine("\\" + oneOf(list(printables)))
reLiteralChar = "".join(c for c in string.printable if c not in r"\[]{}().*?+|")
reRange = Combine(lbrack.suppress() + SkipTo(rbrack,ignore=escapedChar) + rbrack.suppress())
reLiteral = ( escapedChar | oneOf(list(reLiteralChar)) )
reDot = Literal(".")
repetition = (
( lbrace + Word(nums).setResultsName("count") + rbrace ) |
( lbrace + Word(nums).setResultsName("minCount")+","+ Word(nums).setResultsName("maxCount") + rbrace ) |
oneOf(list("*+?"))
)
reExpr = Forward()
reGroup = (lparen.suppress() +
Optional(Literal("?").suppress() + oneOf(list(":P"))).setResultsName("option") +
reExpr.setResultsName("expr") +
rparen.suppress())
reTerm = ( reLiteral | reRange | reMacro | reDot | reGroup )
reExpr << operatorPrecedence( reTerm,
[
(repetition, 1, opAssoc.LEFT, create(Repetition)),
(None, 2, opAssoc.LEFT, create(Sequence)),
(Suppress('|'), 2, opAssoc.LEFT, create(Alternation)),
]
)
reGroup.setParseAction(create(Group))
reRange.setParseAction(create(Range))
reLiteral.setParseAction(create(Character))
reMacro.setParseAction(create(Macro))
reDot.setParseAction(create(Dot))
_parser = reExpr
return _parser
def getkw_bnf(self):
sect_begin = Literal("{").suppress()
sect_end = Literal("}").suppress()
array_begin = Literal("[").suppress()
array_end = Literal("]").suppress()
arg_begin = Literal("(").suppress()
arg_end = Literal(")").suppress()
eql = Literal("=").suppress()
dmark = Literal('$').suppress()
end_data=Literal('$end').suppress()
prtable = alphanums+r'!$%&*+-./<>?@^_|~'
# Helper definitions
kstr=Word(prtable) ^ quotedString.setParseAction(removeQuotes)
name = Word(alphas+"_",alphanums+"_")
vec=array_begin+delimitedList(Word(prtable) ^ \
Literal("\n").suppress() ^ \
quotedString.setParseAction(removeQuotes))+array_end
sect=name+sect_begin
key_sect=name+Group(arg_begin+kstr+arg_end)+sect_begin
vec_sect=name+Group(arg_begin+vec+ arg_end)+sect_begin
# Grammar
keyword = name + eql + kstr
vector = name + eql + vec
data=Combine(dmark+name)+SkipTo(end_data)+end_data
section=Forward()
sect_def=(sect | key_sect | vec_sect)
input=section | data | vector | keyword
section << sect_def+ZeroOrMore(input) + sect_end
# Parsing actions
keyword.setParseAction(self.store_key)
vector.setParseAction(self.store_vector)
data.setParseAction(self.store_data)
sect.setParseAction(self.add_sect)
key_sect.setParseAction(self.add_sect)
vec_sect.setParseAction(self.add_vecsect)
sect_end.setParseAction(self.pop_sect)
bnf=ZeroOrMore(input) + StringEnd().setFailAction(parse_error)
bnf.ignore(pythonStyleComment)
return bnf
def _get_handbrake_title_pattern(self):
title = Literal("+ title").suppress()
integer = Word("0123456789")
time = Combine(integer + ":" + integer + ":" + integer)
duration = Literal("+ duration:").suppress()
subtitle = Literal("+ subtitle tracks:")
iso = Literal('(iso639-2:').suppress() + Word(alphas)
subtitle_track = Literal("+").suppress() + Group(integer + SkipTo(iso).suppress() + iso) + restOfLine.suppress()
title_num = integer.setResultsName("title")
duration_num = time.setResultsName("duration")
subtitles = Group(ZeroOrMore(subtitle_track)).setResultsName("subtitles")
pattern = title + title_num + \
SkipTo(duration).suppress() + \
duration + duration_num + \
SkipTo(subtitle).suppress() + subtitle.suppress() + subtitles
return pattern
def _build_grammar(self):
expr = Forward()
float_lit = Combine(Word(nums) + '.' + Word(nums))
float_lit.setName('float')
float_lit.setParseAction(lambda x: \
self.to_literal(float(x[0])))
int_lit = Word(nums)
int_lit.setName('int')
int_lit.setParseAction(lambda x: \
self.to_literal(int(x[0])))
num = (float_lit | int_lit)
num.setParseAction(lambda x: x[0])
tag_name = Word(alphas + "_", alphanums + "_")
tag_name.setName('tag_name')
tag_name.setParseAction(lambda t: tag_reference.TagReference(t[0]))
quoted_string = QuotedString("'")
quoted_string.setParseAction(lambda s: self.to_literal(s[0]))
oper = oneOf('+ * / -')
oper.setParseAction(lambda o: o[0])
lpar = Literal("(").suppress()
rpar = Literal(")").suppress()
arith = Group(lpar + expr + oper + expr + rpar)
arith.setParseAction(lambda t: \
self.to_arith(t[0][0], t[0][1], t[0][2]))
assign = tag_name + '=' + expr
assign.setName('assign')
assign.setParseAction(lambda x: self.to_assign(x[0],x[2]))
print_tags = Literal('?')
print_tags.setParseAction(lambda x: self.to_print_tags())
expr <<(arith|assign|tag_name|num|quoted_string|print_tags)
expr.setParseAction(lambda x: x[0])
return expr
def instance():
lit_e = CaselessLiteral('E')
plusorminus = Literal('+') | Literal('-')
number = Word(nums)
integer = Combine(Optional(plusorminus) +
number).setParseAction(lambda t:int(t[0]))
index = integer.copy().addParseAction(index_check(0))
floatnumber = Combine( integer +
Optional( Literal('.') + Optional(number) ) +
Optional( lit_e + integer )
).setParseAction(lambda t:float(t[0]))
#comment = Suppress("%") + Word(alphanums + " ")
comment = Regex(r"%.*").setName("comment").suppress()
linend = Or( [comment , LineEnd()] ).suppress()
section_end = (Literal('#') + LineEnd()).suppress()
vertex = (Group( OneOrMore( floatnumber('point') +
OneOrMore( White() ).suppress() ) ) + linend)('vertex')
vertex_header = (Keyword('VERTEX') + linend).suppress()
vertex_section = (vertex_header + Group(OneOrMore(vertex))('vertices') +
section_end)
simplex = (Group( OneOrMore( index('index')
+ OneOrMore( White() ).suppress() ) ) + linend)('simplex')
simplex_header = (Keyword('SIMPLEX') + linend).suppress()
simplex_section = (simplex_header + Group(OneOrMore(simplex))('simplices') +
section_end)
boundarysegment = (Group( index('id') +
OneOrMore( index('index') +
OneOrMore( White() ).suppress() ) ) +
linend)('boundarysegment')
boundarysegment_header = (Keyword('BOUNDARYSEGMENTS') + linend).suppress()
boundarysegment_section = (boundarysegment_header +
Dict(OneOrMore(
boundarysegment ))('boundarysegments') +
section_end)
sections = Each([vertex_section, simplex_section, boundarysegment_section])
dgf_header = (Keyword('DGF') + linend).suppress()
dgf = (dgf_header + Dict(sections) + OneOrMore( section_end ))('dgf')
return dgf
def pattern():
"""pyparsing pattern
"""
def attachLocation(s, loc, tocs):
"""pyparsing callback. Saves path position in the original string
"""
return [(loc, tocs[0])]
path = CharsNotIn(" \t")("path")
path.setParseAction(attachLocation)
longPath = CharsNotIn(" \t", min=2)("path")
longPath.setParseAction(attachLocation)
slashPath = Combine(Literal('/') + Optional(CharsNotIn(" \t")))("path")
slashPath.setParseAction(attachLocation)
pat = ((Literal('f ') + Optional(White()) + Optional(path)) ^ longPath ^ slashPath) + \
Optional(White() + Word(nums)("line"))
pat.leaveWhitespace()
pat.setParseAction(CommandOpen.create)
return pat
def parser():
global _parser
if _parser is None:
ParserElement.setDefaultWhitespaceChars("")
lbrack, rbrack, lbrace, rbrace, lparen, rparen = map(Literal, "[]{}()")
reMacro = Combine("\\" + oneOf(list("dws")))
escapedChar = ~ reMacro + Combine("\\" + oneOf(list(printables)))
reLiteralChar = "".join(c for c in printables if c not in r"\[]{}().*?+|") + " \t"
reRange = Combine(lbrack + SkipTo(rbrack, ignore=escapedChar) + rbrack)
reLiteral = (escapedChar | oneOf(list(reLiteralChar)))
reDot = Literal(".")
repetition = (
(lbrace + Word(nums).setResultsName("count") + rbrace) |
(lbrace + Word(nums).setResultsName("minCount") + "," + Word(nums).setResultsName("maxCount") + rbrace) |
oneOf(list("*+?"))
)
reRange.setParseAction(handle_range)
reLiteral.setParseAction(handle_literal)
reMacro.setParseAction(handle_macro)
reDot.setParseAction(handle_dot)
reTerm = (reLiteral | reRange | reMacro | reDot)
reExpr = operatorPrecedence(reTerm, [
(repetition, 1, opAssoc.LEFT, handle_repetition),
(None, 2, opAssoc.LEFT, handle_sequence),
(Suppress('|'), 2, opAssoc.LEFT, handle_alternative),
])
_parser = reExpr
return _parser
def getLogLineBNF():
global logLineBNF
if logLineBNF is None:
integer = Word( nums )
ipAddress = delimitedList( integer, ".", combine=True )
timeZoneOffset = Word("+-",nums)
month = Word(string.uppercase, string.lowercase, exact=3)
serverDateTime = Combine( Suppress("[") +
Combine( integer + "/" + month + "/" + integer +
":" + integer + ":" + integer + ":" + integer + ' ' + timeZoneOffset) +
Suppress("]") )
logLineBNF = ( ipAddress.setResultsName("ipAddr") +
Suppress("-") +
("-" | Word( alphas+nums+"@._" )).setResultsName("auth") +
serverDateTime.setResultsName("timestamp") +
dblQuotedString.setResultsName("cmd").setParseAction(getCmdFields) +
(integer | "-").setResultsName("statusCode") +
(integer | "-").setResultsName("numBytesSent") +
dblQuotedString.setResultsName("referrer").setParseAction(removeQuotes) +
dblQuotedString.setResultsName("clientSfw").setParseAction(removeQuotes) )
return logLineBNF
def urlsplit(url, scheme='', allow_fragments=1):
"""Parse a URL into 5 components:
<scheme>://<netloc>/<path>?<query>#<fragment>
Return a 5-tuple: (scheme, netloc, path, query, fragment).
Note that we don't break the components up in smaller bits
(e.g. netloc is a single string) and we don't expand % escapes."""
global _urlBNF
key = url, scheme, allow_fragments
cached = _parse_cache.get(key, None)
if cached:
return cached
if len(_parse_cache) >= MAX_CACHE_SIZE: # avoid runaway growth
clear_cache()
if (_urlBNF is None):
scheme_chars = alphanums + "+-."
urlscheme = Word( scheme_chars )
netloc_chars = "".join( [ c for c in printables if c not in "/." ] )
netloc = Combine(delimitedList( Word( netloc_chars ), ".", combine=True ))
path_chars = "".join( [ c for c in printables if c not in "?" ] )
path = Word( path_chars )
query_chars = "".join( [ c for c in printables if c not in "#" ] )
query = Word( query_chars )
fragment = Word( printables+" " )
_urlBNF = Combine(Optional(urlscheme.setResultsName("scheme") + ":" ) +
Optional(Literal("//").suppress() + netloc, default="").setResultsName("netloc") +
Optional(path.setResultsName("path"), default="") +
Optional(Literal("?").suppress() + query, default="").setResultsName("query") +
Optional(Literal("#").suppress() + fragment, default="").setResultsName("fragment") )
tokens = _urlBNF.parseString( url )
tuple = (tokens.scheme or scheme), tokens.netloc[0], tokens.path, tokens.query[0], tokens.fragment[0]
_parse_cache[key] = tuple
return tuple
def _define_grammar(self):
'''define the grammar to be used, and add actions'''
self._define_actions()
eol = LineEnd().suppress()
white = Optional(White()).suppress()
begin = Keyword('begin').suppress()
end = Keyword('end').suppress()
comment = (Literal('#') + restOfLine).suppress()
data_value = Combine(OneOrMore(CharsNotIn('#\n\r')))
data_line = (LineStart() + white + Optional(data_value) +
Optional(comment) + eol)
block_name = Word(alphas, alphanums + '_')
begin_block = (LineStart() + begin + block_name +
Optional(comment) + eol)
end_block = LineStart() + end + block_name + Optional(comment) + eol
junk = ZeroOrMore(LineStart() + white + NotAny(begin) +
restOfLine + eol).suppress()
data = Group(ZeroOrMore(NotAny(end) + data_line))
block_def = begin_block + data + end_block
block_defs = junk + OneOrMore(block_def + junk)
self._grammar = block_defs
begin_block.addParseAction(self._begin_block_action)
end_block.addParseAction(self._end_block_action)
data_value.addParseAction(self._data_value_action)
def grammar(self):
number = Combine(Word(nums) + Optional("." + OneOrMore(Word(nums))))
table = Combine(Word(alphas) + OneOrMore(Word("_"+alphanums)))
number.setParseAction(self._number_parse_action)
table.setParseAction(self._table_parse_action)
signop = oneOf('+ -')
multop = oneOf('* /')
plusop = oneOf('+ -')
operand = number | table
return operatorPrecedence(operand, [
(signop, 1, opAssoc.RIGHT),
(multop, 2, opAssoc.LEFT),
(plusop, 2, opAssoc.LEFT)
])
def create_bnf():
cvt_int = lambda toks: int(toks[0])
cvt_real = lambda toks: float(toks[0])
cvt_tuple = lambda toks : tuple(toks.asList())
cvt_dict = lambda toks: dict(toks.asList())
# define punctuation as suppressed literals
(lparen, rparen, lbrack, rbrack,
lbrace, rbrace, colon) = map(Suppress,"()[]{}:")
integer = Combine(Optional(oneOf("+ -")) + Word(nums)).setName("integer")
integer.setParseAction(cvt_int)
real = Combine(Optional(oneOf("+ -"))+ Word(nums)
+ "." + Optional(Word(nums))
+ Optional("e" + Optional(oneOf("+ -"))
+ Word(nums))).setName("real")
real.setParseAction(cvt_real)
tuple_str = Forward()
list_str = Forward()
dict_str = Forward()
list_item = (real | integer | Group(list_str) | tuple_str | dict_str
| quotedString.setParseAction(removeQuotes)
| Word(alphas8bit + alphas, alphas8bit + alphanums + "_"))
list_item2 = list_item | Empty().setParseAction(lambda: [None])
tuple_str << (Suppress("(") + Optional(delimitedList(list_item)) +
Optional(Suppress(",")) + Suppress(")"))
tuple_str.setParseAction(cvt_tuple)
list_str << (lbrack + Optional(delimitedList(list_item) +
Optional(Suppress(","))) + rbrack)
dict_entry = Group(list_item + colon + list_item2)
dict_inner = delimitedList(dict_entry) + Optional(Suppress(","))
dict_inner.setParseAction(cvt_dict)
dict_str << (lbrace + Optional(dict_inner) + rbrace)
return dict_inner
def parser():
global _parser
if _parser is None:
ParserElement.setDefaultWhitespaceChars("")
lbrack,rbrack,lbrace,rbrace,lparen,rparen,colon,qmark = map(Literal,"[]{}():?")
reMacro = Combine("\\" + oneOf(list("dws")))
escapedChar = ~reMacro + Combine("\\" + oneOf(list(printables)))
reLiteralChar = "".join(c for c in printables if c not in r"\[]{}().*?+|") + " \t"
reRange = Combine(lbrack + SkipTo(rbrack,ignore=escapedChar) + rbrack)
reLiteral = ( escapedChar | oneOf(list(reLiteralChar)) )
reNonCaptureGroup = Suppress("?:")
reDot = Literal(".")
repetition = (
( lbrace + Word(nums)("count") + rbrace ) |
( lbrace + Word(nums)("minCount")+","+ Word(nums)("maxCount") + rbrace ) |
oneOf(list("*+?"))
)
reRange.setParseAction(handleRange)
reLiteral.setParseAction(handleLiteral)
reMacro.setParseAction(handleMacro)
reDot.setParseAction(handleDot)
reTerm = ( reLiteral | reRange | reMacro | reDot | reNonCaptureGroup)
reExpr = infixNotation( reTerm,
[
(repetition, 1, opAssoc.LEFT, handleRepetition),
(None, 2, opAssoc.LEFT, handleSequence),
(Suppress('|'), 2, opAssoc.LEFT, handleAlternative),
]
)
_parser = reExpr
return _parser
See http://pyparsing.wikispaces.com/message/view/home/620225
and the discussion of the "-" operator in the docs.
'''
ParsingTmp.keywords.append(x)
W = Where
O = Optional
S = Suppress
number = Word(nums)
point = Literal('.')
e = CaselessLiteral('E')
plusorminus = Literal('+') | Literal('-')
integer = Combine(O(plusorminus) + number)
floatnumber = Combine(integer + (point + O(number)) ^ (e + integer))
integer.setParseAction(lambda tokens: SimpleRValue(int(tokens[0])))
floatnumber.setParseAction(lambda tokens: SimpleRValue(float(tokens[0])))
pi = Keyword('pi').setParseAction(lambda tokens: SimpleRValue(math.pi, 'pi')) #@UnusedVariable
isnumber = lambda x: isinstance(x, Number)
rvalue = Forward()
rvalue.setName('rvalue')
contract_expression = Forward()
contract_expression.setName('contract')
simple_contract = Forward()
simple_contract.setName('simple_contract')
# Import all expressions -- they will call add_contract()
from .library import (EqualTo, Unary, Binary, composite_contract,
def get_grammar(self):
"""
Defines our grammar for mathematical expressions.
Possibly helpful:
- BNF form of context-free grammar https://en.wikipedia.org/wiki/Backus%E2%80%93Naur_form
- Some pyparsing docs http://infohost.nmt.edu/~shipman/soft/pyparsing/web/index.html
"""
# Define + and -
plus = Literal("+")
minus = Literal("-")
plus_minus = plus | minus
# 1 or 1.0 or .1
number_part = Word(nums)
inner_number = Combine((number_part +
Optional("." + Optional(number_part)))
| ("." + number_part))
# Combine() joints the matching parts together in a single token,
# and requires that the matching parts be contiguous (no spaces)
# Define our suffixes
suffix = Word(alphas + '%')
suffix.setParseAction(self.suffix_parse_action)
# Construct number as a group consisting of a text string ("num") and an optional suffix.
# num can include a decimal number and numerical exponent, and can be
# converted to a number using float()
# suffix may contain alphas or %
# Spaces are ignored inside numbers
# Group wraps everything up into its own ParseResults object when parsing
number = Group(
Combine(
inner_number + Optional(
CaselessLiteral("E") + Optional(plus_minus) +
number_part), )("num") +
Optional(suffix)("suffix"))("number")
# Note that calling ("name") on the end of a parser is equivalent to calling
# parser.setResultsName, which is used to pull that result out of a parsed
# expression like a dictionary.
# Construct variable and function names
front = Word(alphas, alphanums) # must start with alpha
subscripts = Word(alphanums + '_') + ~FollowedBy('{') # ~ = not
lower_indices = Literal("_{") + Optional("-") + Word(
alphanums) + Literal("}")
upper_indices = Literal("^{") + Optional("-") + Word(
alphanums) + Literal("}")
# Construct an object name in either of two forms:
# 1. front + subscripts + tail
# 2. front + lower_indices + upper_indices + tail
# where:
# front (required):
# starts with alpha, followed by alphanumeric
# subscripts (optional):
# any combination of alphanumeric and underscores
# lower_indices (optional):
# Of form "_{(-)<alphanumeric>}"
# upper_indices (optional):
# Of form "^{(-)<alphanumeric>}"
# tail (optional):
# any number of primes
name = Combine(front +
Optional(subscripts | (Optional(lower_indices) +
Optional(upper_indices))) +
ZeroOrMore("'"))
# Define a variable as a pyparsing result that contains one object name
variable = Group(name("varname"))("variable")
variable.setParseAction(self.variable_parse_action)
# initialize recursive grammar
expression = Forward()
# Construct functions as consisting of funcname and arguments as
# funcname(arguments)
# where arguments is a comma-separated list of arguments, returned as a list
# Must have at least 1 argument
function = Group(
name("funcname") + Suppress("(") +
Group(delimitedList(expression))("arguments") +
Suppress(")"))("function")
function.setParseAction(self.function_parse_action)
# Define parentheses
parentheses = Group(Suppress("(") + expression +
Suppress(")"))('parentheses')
# Define arrays
array = Group(
Suppress("[") + delimitedList(expression) + Suppress("]"))("array")
# atomic units evaluate directly to number or array without binary operations
atom = number | function | variable | parentheses | array
# Define operations in order of precedence
# Define exponentiation, possibly including negative powers
power = atom + ZeroOrMore(Suppress("^") + Optional(minus)("op") + atom)
power.addParseAction(self.group_if_multiple('power'))
# Define negation (e.g., in 5*-3 --> we need to evaluate the -3 first)
# Negation in powers is handled separately
# This has been arbitrarily assigned a higher precedence than parallel
negation = Optional(minus)("op") + power
negation.addParseAction(self.group_if_multiple('negation'))
# Define the parallel operator 1 || 5 == 1/(1/1 + 1/5)
pipes = Literal('|') + Literal('|')
parallel = negation + ZeroOrMore(Suppress(pipes) + negation)
parallel.addParseAction(self.group_if_multiple('parallel'))
# Define multiplication and division
product = parallel + ZeroOrMore((Literal('*') | Literal('/'))("op") +
parallel)
product.addParseAction(self.group_if_multiple('product'))
# Define sums and differences
# Note that leading - signs are treated by negation
sumdiff = Optional(plus) + product + ZeroOrMore(
plus_minus("op") + product)
sumdiff.addParseAction(self.group_if_multiple('sum'))
# Close the recursion
expression << sumdiff
return expression + stringEnd
PERIOD = Literal(".").suppress()
RANGLE = Literal(">").suppress()
RBRACE = Literal("]").suppress()
RPAREN = Literal(")").suppress()
CATEGORIES = CaselessLiteral("categories").suppress()
END = CaselessLiteral("end").suppress()
FONT = CaselessLiteral("font").suppress()
HINT = CaselessLiteral("hint").suppress()
ITEM = CaselessLiteral("item").suppress()
OBJECT = CaselessLiteral("object").suppress()
attribute_value_pair = Forward() # this is recursed in item_list_entry
simple_identifier = Word(alphas, alphanums + "_")
identifier = Combine( simple_identifier + ZeroOrMore( Literal(".") + simple_identifier ))
object_name = identifier
object_type = identifier
# Integer and floating point values are converted to Python longs and floats, respectively.
int_value = Combine(Optional("-") + Word(nums)).setParseAction(lambda s,l,t: [ int(t[0]) ] )
float_value = Combine(Optional("-") + Optional(Word(nums)) + "." + Word(nums)).setParseAction(lambda s,l,t: [ float(t[0]) ] )
number_value = float_value | int_value
# Base16 constants are left in string form, including the surrounding braces.
base16_value = Combine(Literal("{") + OneOrMore(Word("0123456789ABCDEFabcdef")) + Literal("}"), adjacent=False)
# This is the first part of a hack to convert the various delphi partial sglQuotedStrings
# into a single sglQuotedString equivalent. The gist of it is to combine
# all sglQuotedStrings (with their surrounding quotes removed (suppressed))
# with sequences of #xyz character constants, with "strings" concatenated
def graph_definition():
global graphparser
if not graphparser:
# punctuation
colon = Literal(":")
lbrace = Literal("{")
rbrace = Literal("}")
lbrack = Literal("[")
rbrack = Literal("]")
lparen = Literal("(")
rparen = Literal(")")
equals = Literal("=")
comma = Literal(",")
dot = Literal(".")
slash = Literal("/")
bslash = Literal("\\")
star = Literal("*")
semi = Literal(";")
at = Literal("@")
minus = Literal("-")
# keywords
strict_ = Literal("strict")
graph_ = Literal("graph")
digraph_ = Literal("digraph")
subgraph_ = Literal("subgraph")
node_ = Literal("node")
edge_ = Literal("edge")
# token definitions
identifier = Word(alphanums + "_").setName("identifier")
double_quoted_string = dblQuotedString
alphastring_ = OneOrMore(CharsNotIn(_noncomma))
ID = (identifier | double_quoted_string.setParseAction(strip_quotes) |\
alphastring_).setName("ID")
html_text = Combine(Literal("<<") + OneOrMore(CharsNotIn(",]")))
float_number = Combine(Optional(minus) + \
OneOrMore(Word(nums + "."))).setName("float_number")
righthand_id = (float_number | ID | html_text).setName("righthand_id")
port_angle = (at + ID).setName("port_angle")
port_location = (Group(colon + ID) | \
Group(colon + lparen + ID + comma + ID + rparen)).setName("port_location")
port = (Group(port_location + Optional(port_angle)) | \
Group(port_angle + Optional(port_location))).setName("port")
node_id = (ID + Optional(port))
a_list = OneOrMore(ID + Optional(equals.suppress() + righthand_id) + \
Optional(comma.suppress())).setName("a_list")
attr_list = OneOrMore(lbrack.suppress() + Optional(a_list) + \
rbrack.suppress()).setName("attr_list")
attr_stmt = (Group(graph_ | node_ | edge_) +
attr_list).setName("attr_stmt")
edgeop = (Literal("--") | Literal("->")).setName("edgeop")
stmt_list = Forward()
graph_stmt = Group(lbrace.suppress() + Optional(stmt_list) + \
rbrace.suppress()).setName("graph_stmt")
subgraph = (Group(Optional(subgraph_ + Optional(ID)) + graph_stmt) | \
Group(subgraph_ + ID)).setName("subgraph")
edgeRHS = OneOrMore(edgeop + Group(node_id | subgraph))
edge_stmt = Group(node_id | subgraph) + edgeRHS + Optional(attr_list)
node_stmt = (node_id + Optional(attr_list) +
Optional(semi.suppress())).setName("node_stmt")
assignment = (ID + equals.suppress() +
righthand_id).setName("assignment")
stmt = (assignment | edge_stmt | attr_stmt | subgraph
| node_stmt).setName("stmt")
stmt_list << OneOrMore(stmt + Optional(semi.suppress()))
graphparser = (Optional(strict_) + Group((graph_ | digraph_)) + \
Optional(ID) + graph_stmt).setResultsName("graph")
singleLineComment = "//" + restOfLine
# actions
graphparser.ignore(singleLineComment)
graphparser.ignore(cStyleComment)
assignment.setParseAction(push_attr_list)
a_list.setParseAction(push_attr_list)
edge_stmt.setParseAction(push_edge_stmt)
node_stmt.setParseAction(push_node_stmt)
attr_stmt.setParseAction(push_default_stmt)
subgraph.setParseAction(push_subgraph_stmt)
graph_stmt.setParseAction(push_graph_stmt)
graphparser.setParseAction(push_top_graph_stmt)
return graphparser
return f"{node.fn}({', '.join(pprint(a) for a in node.args)})"
elif isinstance(node, Nary):
return "({})".format(f" {node.fn} ".join(pprint(a) for a in node.args))
else:
return node
EXPR = Forward()
BEXPR = Forward()
OFFSET = LBRAK + EXPR + RBRAK
EXPRATOM = (
ppc.signed_integer |
(VARNAME + Optional(OFFSET, default=None) + Keyword("of").suppress() +
VARNAME).setParseAction(lambda toks: OfNode(*toks)) | # noqa: E501
(Combine(BUILTIN + LPAR) + Group(delimitedList(EXPR)) +
RPAR).setParseAction(lambda toks: BuiltIn(*toks)) # noqa: E501
)
EXPR <<= infixNotation(
EXPRATOM,
[("%", 2, opAssoc.LEFT, lambda x: BinOp(*x[0])),
(oneOf("* /"), 2, opAssoc.LEFT, lambda x: BinOp(*x[0])),
(oneOf("+ - "), 2, opAssoc.LEFT, lambda x: BinOp(*x[0])),
(oneOf("> < = >= <= !="), 2, opAssoc.LEFT, lambda x: BinOp(*x[0]))])
BEXPR <<= infixNotation(
EXPR,
[
# Note: "not" is implemented as a BuiltIn
(oneOf("and or"), 2, opAssoc.LEFT, lambda x: BinOp(*x[0]))
def convertToFloat(s, loc, toks):
try:
return float(toks[0])
except:
raise ParseException(loc, "invalid float format %s"%toks[0])
exponent = CaselessLiteral("e")+Optional(sign)+Word(nums)
#note that almost all these fields are optional,
#and this can match almost anything. We rely on Pythons built-in
#float() function to clear out invalid values - loosely matching like this
#speeds up parsing quite a lot
floatingPointConstant = Combine(
Optional(sign) +
Optional(Word(nums)) +
Optional(Literal(".") + Optional(Word(nums)))+
Optional(exponent)
)
floatingPointConstant.setParseAction(convertToFloat)
number = floatingPointConstant
#same as FP constant but don't allow a - sign
nonnegativeNumber = Combine(
Optional(Word(nums)) +
Optional(Literal(".") + Optional(Word(nums)))+
Optional(exponent)
)
nonnegativeNumber.setParseAction(convertToFloat)
(even binary content) inside the structure. This is done by pre-
sizing the data with the NUMBER similar to Dan Bernstein's netstrings
setup.
SPACE White space is basically ignored. This is interesting because since
Stackish is serialized consistently this means you can use \n as the
separation character and perform reasonable diffs on two structures.
"""
from pyparsing import Suppress,Word,nums,alphas,alphanums,Combine,oneOf,\
Optional,QuotedString,Forward,Group,ZeroOrMore,srange
MARK, UNMARK, AT, COLON, QUOTE = map(Suppress, "[]@:'")
NUMBER = Word(nums)
NUMBER.setParseAction(lambda t: int(t[0]))
FLOAT = Combine(oneOf("+ -") + Word(nums) + "." + Optional(Word(nums)))
FLOAT.setParseAction(lambda t: float(t[0]))
STRING = QuotedString('"', multiline=True)
WORD = Word(alphas, alphanums + "_:")
ATTRIBUTE = Combine(AT + WORD)
strBody = Forward()
def setBodyLength(tokens):
strBody << Word(srange(r'[\0x00-\0xffff]'), exact=int(tokens[0]))
return ""
BLOB = Combine(QUOTE + Word(nums).setParseAction(setBodyLength) + COLON +
strBody + QUOTE)
from pyparsing import alphas, alphanums, Combine, delimitedList, Forward, Group, Literal, \
Keyword, nums, oneOf, Optional, ParserElement, Suppress, White, Word
ParserElement.enablePackrat()
LPAR, RPAR = map(Suppress, '()')
const = Literal('true') | Literal('false')
AOps = oneOf('INTS_MODULUS_TOTAL * / + -').setParseAction(
lambda s, l, t: ['%'] if t[0] == 'INTS_MODULUS_TOTAL' else t)
BOps = (Keyword('and').setParseAction(lambda s, l, t: ['&'])
| Keyword('not').setParseAction(lambda s, l, t: ['!'])
| Keyword('or').setParseAction(lambda s, l, t: ['|']))
ROps = oneOf('< > <= >= =')
val = Combine(Optional('-') + Word(nums))
var = Word(alphas + '_:$', alphanums + '_:$')
term = val | var
let = Forward()
pred = Forward()
stmt = Forward()
expr = Forward()
expr << (term
|
(LPAR + AOps + Group(delimitedList(expr, delim=White(' '))) + RPAR
).setParseAction(lambda s, l, t: [list(joinit(t[1], t[0]))] if not (
t[0] == '-' and len(t[1]) == 1) else [['0 -', t[1][0]]])
| (LPAR + expr + RPAR))
See http://pyparsing.wikispaces.com/message/view/home/620225
and the discussion of the "-" operator in the docs.
'''
ParsingTmp.keywords.append(x)
W = Where
O = Optional
S = Suppress
number = Word(nums)
point = Literal('.')
e = CaselessLiteral('E')
plusorminus = Literal('+') | Literal('-')
integer = Combine(O(plusorminus) + number)
# warnings.warn('TODO: negative float number')
floatnumber = Combine(integer + (point + O(number)) ^ (e + integer))
integer.setParseAction(lambda tokens: SimpleRValue(int(tokens[0])))
floatnumber.setParseAction(lambda tokens: SimpleRValue(float(tokens[0])))
pi = Keyword('pi').setParseAction(
lambda tokens: SimpleRValue(math.pi, 'pi')) # @UnusedVariable
def isnumber(x):
# These are scalar quantities that we can compare (=,>,>=, etc.)
if isinstance(x, Number):
return True
try:
# Slow, do it only once (TODO)
import numpy
ident = Word(alphas, alphanums + "_$").setName("identifier")
columnName = Upcase(delimitedList(ident, ".", combine=True))
columnNameList = Group(delimitedList(columnName))
tableName = Upcase(delimitedList(ident, ".", combine=True))
tableNameList = Group(delimitedList(tableName))
whereExpression = Forward()
and_ = Keyword("and", caseless=True)
or_ = Keyword("or", caseless=True)
in_ = Keyword("in", caseless=True)
E = CaselessLiteral("E")
binop = oneOf("= != < > >= <= eq ne lt le gt ge", caseless=True)
arithSign = Word("+-", exact=1)
realNum = Combine(
Optional(arithSign) +
(Word(nums) + "." + Optional(Word(nums)) | ("." + Word(nums))) +
Optional(E + Optional(arithSign) + Word(nums)))
intNum = Combine(
Optional(arithSign) + Word(nums) +
Optional(E + Optional("+") + Word(nums)))
columnRval = realNum | intNum | quotedString | columnName # need to add support for alg expressions
whereCondition = Group((columnName + binop + columnRval)
| (columnName + in_ + "(" + delimitedList(columnRval) +
")") | (columnName + in_ + "(" + selectStmt + ")")
| ("(" + whereExpression + ")"))
whereExpression << whereCondition + ZeroOrMore((and_ | or_) + whereExpression)
# define the grammar
selectStmt << (selectToken + ('*' | columnNameList).setResultsName("columns") +
fromToken + tableNameList.setResultsName("tables") +
def graph_definition():
global graphparser
if not graphparser:
# punctuation
colon = Literal(":")
lbrace = Literal("{")
rbrace = Literal("}")
lbrack = Literal("[")
rbrack = Literal("]")
lparen = Literal("(")
rparen = Literal(")")
equals = Literal("=")
comma = Literal(",")
dot = Literal(".")
slash = Literal("/")
bslash = Literal("\\")
star = Literal("*")
semi = Literal(";")
at = Literal("@")
minus = Literal("-")
# keywords
strict_ = CaselessLiteral("strict")
graph_ = CaselessLiteral("graph")
digraph_ = CaselessLiteral("digraph")
subgraph_ = CaselessLiteral("subgraph")
node_ = CaselessLiteral("node")
edge_ = CaselessLiteral("edge")
# token definitions
identifier = Word(alphanums + "_." ).setName("identifier")
double_quoted_string = QuotedString('"', escChar="\\", multiline=True, unquoteResults=False) # dblQuotedString
_noncomma = "".join( [ c for c in printables if c != "," ] )
alphastring_ = OneOrMore(CharsNotIn(_noncomma + ' '))
def parse_html(s, loc, toks):
return '<%s>' % ''.join(toks[0])
opener = '<'
closer = '>'
html_text = nestedExpr( opener, closer,
( CharsNotIn( opener + closer ) )
).setParseAction(parse_html).leaveWhitespace()
ID = ( identifier | html_text |
double_quoted_string | #.setParseAction(strip_quotes) |
alphastring_ ).setName("ID")
float_number = Combine(Optional(minus) +
OneOrMore(Word(nums + "."))).setName("float_number")
righthand_id = (float_number | ID ).setName("righthand_id")
port_angle = (at + ID).setName("port_angle")
port_location = (OneOrMore(Group(colon + ID)) |
Group(colon + lparen + ID + comma + ID + rparen)).setName("port_location")
port = (Group(port_location + Optional(port_angle)) |
Group(port_angle + Optional(port_location))).setName("port")
node_id = (ID + Optional(port))
a_list = OneOrMore(ID + Optional(equals + righthand_id) +
Optional(comma.suppress())).setName("a_list")
attr_list = OneOrMore(lbrack.suppress() + Optional(a_list) +
rbrack.suppress()).setName("attr_list")
attr_stmt = (Group(graph_ | node_ | edge_) + attr_list).setName("attr_stmt")
edgeop = (Literal("--") | Literal("->")).setName("edgeop")
stmt_list = Forward()
graph_stmt = Group(lbrace.suppress() + Optional(stmt_list) +
rbrace.suppress() + Optional(semi.suppress()) ).setName("graph_stmt")
edge_point = Forward()
edgeRHS = OneOrMore(edgeop + edge_point)
edge_stmt = edge_point + edgeRHS + Optional(attr_list)
subgraph = Group(subgraph_ + Optional(ID) + graph_stmt).setName("subgraph")
edge_point << Group( subgraph | graph_stmt | node_id ).setName('edge_point')
node_stmt = (node_id + Optional(attr_list) + Optional(semi.suppress())).setName("node_stmt")
assignment = (ID + equals + righthand_id).setName("assignment")
stmt = (assignment | edge_stmt | attr_stmt | subgraph | graph_stmt | node_stmt).setName("stmt")
stmt_list << OneOrMore(stmt + Optional(semi.suppress()))
graphparser = OneOrMore( (Optional(strict_) + Group((graph_ | digraph_)) +
Optional(ID) + graph_stmt).setResultsName("graph") )
singleLineComment = Group("//" + restOfLine) | Group("#" + restOfLine)
# actions
graphparser.ignore(singleLineComment)
graphparser.ignore(cStyleComment)
assignment.setParseAction(push_attr_list)
a_list.setParseAction(push_attr_list)
edge_stmt.setParseAction(push_edge_stmt)
node_stmt.setParseAction(push_node_stmt)
attr_stmt.setParseAction(push_default_stmt)
subgraph.setParseAction(push_subgraph_stmt)
graph_stmt.setParseAction(push_graph_stmt)
graphparser.setParseAction(push_top_graph_stmt)
return graphparser
def pushFirst(str, loc, toks):
exprStack.append(toks[0])
def assignVar(str, loc, toks):
varStack.append(toks[0])
# define grammar
point = Literal('.')
e = CaselessLiteral('E')
plusorminus = Literal('+') | Literal('-')
number = Word(nums)
integer = Combine(Optional(plusorminus) + number)
floatnumber = Combine(integer + Optional(point + Optional(number)) +
Optional(e + integer))
ident = Word(alphas, alphanums + '_')
plus = Literal("+")
minus = Literal("-")
mult = Literal("*")
div = Literal("/")
lpar = Literal("(").suppress()
rpar = Literal(")").suppress()
addop = plus | minus
multop = mult | div
expop = Literal("^")
assign = Literal("=")
E = CaselessLiteral("E")
EQ = '='
LOGICAL_OPS = ['and', 'or']
NE = '!='
NE_OPS = [NE, '!~', '>', '<']
NOT = 'not'
PARENTHESES = ['(', ')']
BINOP = \
oneOf(
"== %s =? =^ =$ =~ !~ intersects >= > <= < before after" % NE,
caseless=True)
ARITH_SIGN = Word("+-", exact=1)
REAL_NUM = \
Combine(
Optional(ARITH_SIGN) +
(Word(nums) + "." + Optional(Word(nums)) | ("." + Word(nums))) +
Optional(E + Optional(ARITH_SIGN) + Word(nums)))
INT_NUM = \
Combine(
Optional(ARITH_SIGN) + Word(nums) +
Optional(E + Optional("+") + Word(nums)))
R_VALUE = REAL_NUM | INT_NUM | quotedString | FIELD_NAME
OXTL_EXPRESSION = Forward()
OXTL_CLAUSE = Group((FIELD_NAME + BINOP + R_VALUE)
| (FIELD_NAME + IN_ + PARENTHESES[0] +
delimitedList(R_VALUE) + PARENTHESES[1])
| (FIELD_NAME + IN_ + PARENTHESES[0] + Forward() +
PARENTHESES[1])
| (PARENTHESES[0] + OXTL_EXPRESSION + PARENTHESES[1])
val2 = _eval(archive, context, app, exp_context)
val1 = or_(val1, val2)
return val1
class EvalGroupOp(object):
def __init__(self, tokens):
self._evals = [t.eval for t in tokens[0][0::2]]
def eval(self, archive, context, app, exp_context):
val = [eval(archive, context, app, exp_context) for eval in self._evals]
return val
integer = Word(nums)
real = Combine(Word(nums) + '.' + Word(nums))
constant = (Literal('True') |
Literal('False') |
Literal('None') |
Literal('yes') |
Literal('no')) + WordEnd()
model_reference = Regex(r'([\w\.]*#[\w\.]+)')
variable = Regex(r'([a-zA-Z0-9\._]+)')
string = QuotedString('"', escChar="\\") | QuotedString('\'', escChar="\\")
operand = model_reference | real | integer | constant | string | variable
plusop = oneOf('+ -')
multop = oneOf('* / // %')
else:
query &= t
return query
NO_BRTS = printables.replace('(', '').replace(')', '')
SINGLE = Word(NO_BRTS.replace('*', ''))
WILDCARDS = Optional('*') + SINGLE + Optional('*') + WordEnd(wordChars=NO_BRTS)
QUOTED = quotedString.setParseAction(removeQuotes)
OPER_AND = CaselessLiteral('and')
OPER_OR = CaselessLiteral('or')
OPER_NOT = '-'
TERM = Combine(Optional(Word(alphas).setResultsName('meta') + ':') +
(QUOTED.setResultsName('query') |
WILDCARDS.setResultsName('query')))
TERM.setParseAction(createQ)
EXPRESSION = operatorPrecedence(TERM, [
(OPER_NOT, 1, opAssoc.RIGHT),
(OPER_OR, 2, opAssoc.LEFT),
(Optional(OPER_AND, default='and'), 2, opAssoc.LEFT)])
EXPRESSION.setParseAction(unionQ)
QUERY = OneOrMore(EXPRESSION) + StringEnd()
QUERY.setParseAction(unionQ)
def advanced_search(pattern):
"""Parse the grammar of a pattern
#===============================================================================
FREE_TEXT = Word(printables + ' ', excludeChars='()')
INTEGER = Word(nums)
ID_TEXT = Word(alphanums, alphanums + ':/_-.')
ONTOLOGY_SUFFIX = (Keyword('ABI')
| Keyword('FM')
| Keyword('FMA')
| Keyword('ILX')
| Keyword('MA')
| Keyword('NCBITaxon')
| Keyword('UBERON'))
ONTOLOGY_ID = Combine(ONTOLOGY_SUFFIX + ':' + ID_TEXT)
#===============================================================================
IDENTIFIER = Group(Keyword('id') + Suppress('(') + ID_TEXT + Suppress(')'))
MODELS = Group(Keyword('models') + Suppress('(') + ONTOLOGY_ID + Suppress(')'))
ZOOM_LEVEL = INTEGER
ZOOM = Group(
Keyword('zoom') + Suppress('(') +
Group(ZOOM_LEVEL + Suppress(',') + ZOOM_LEVEL + Suppress(',') +
ZOOM_LEVEL) + Suppress(')'))
LAYER_DIRECTIVES = IDENTIFIER | MODELS | ZOOM
LAYER_DIRECTIVE = '.' + ZeroOrMore(LAYER_DIRECTIVES)
class NginxParser(object):
# pylint: disable=expression-not-assigned
"""A class that parses nginx configuration with pyparsing."""
# constants
space = Optional(White())
nonspace = Regex(r"\S+")
left_bracket = Literal("{").suppress()
right_bracket = space.leaveWhitespace() + Literal("}").suppress()
semicolon = Literal(";").suppress()
key = Word(alphanums + "_/+-.")
dollar_var = Combine(Literal('$') + Regex(r"[^\{\};,\s]+"))
condition = Regex(r"\(.+\)")
# Matches anything that is not a special character, and ${SHELL_VARS}, AND
# any chars in single or double quotes
# All of these COULD be upgraded to something like
# https://stackoverflow.com/a/16130746
dquoted = Regex(r'(\".*\")')
squoted = Regex(r"(\'.*\')")
nonspecial = Regex(r"[^\{\};,]")
varsub = Regex(r"(\$\{\w+\})")
# nonspecial nibbles one character at a time, but the other objects take
# precedence. We use ZeroOrMore to allow entries like "break ;" to be
# parsed as assignments
value = Combine(ZeroOrMore(dquoted | squoted | varsub | nonspecial))
location = CharsNotIn("{};," + string.whitespace)
# modifier for location uri [ = | ~ | ~* | ^~ ]
modifier = Literal("=") | Literal("~*") | Literal("~") | Literal("^~")
# rules
comment = space + Literal('#') + restOfLine
assignment = space + key + Optional(space + value,
default=None) + semicolon
location_statement = space + Optional(modifier) + Optional(space +
location +
space)
if_statement = space + Literal("if") + space + condition + space
charset_map_statement = space + Literal(
"charset_map") + space + value + space + value
map_statement = space + Literal(
"map") + space + nonspace + space + dollar_var + space
# This is NOT an accurate way to parse nginx map entries; it's almost
# certainly too permissive and may be wrong in other ways, but it should
# preserve things correctly in mmmmost or all cases.
#
# - I can neither prove nor disprove that it is correct wrt all escaped
# semicolon situations
# Addresses https://github.com/fatiherikli/nginxparser/issues/19
map_pattern = Regex(r'".*"') | Regex(r"'.*'") | nonspace
map_entry = space + map_pattern + space + value + space + semicolon
map_block = Group(
Group(map_statement).leaveWhitespace() + left_bracket +
Group(ZeroOrMore(Group(comment | map_entry)) +
space).leaveWhitespace() + right_bracket)
block = Forward()
# key could for instance be "server" or "http", or "location" (in which case
# location_statement needs to have a non-empty location)
block_begin = (Group(space + key + location_statement)
^ Group(if_statement)
^ Group(charset_map_statement)).leaveWhitespace()
block_innards = Group(
ZeroOrMore(Group(comment | assignment) | block | map_block) +
space).leaveWhitespace()
block << Group(block_begin + left_bracket + block_innards + right_bracket)
script = OneOrMore(Group(comment | assignment) ^ block
^ map_block) + space + stringEnd
script.parseWithTabs().leaveWhitespace()
def __init__(self, source):
self.source = source
def parse(self):
"""Returns the parsed tree."""
return self.script.parseString(self.source)
def as_list(self):
"""Returns the parsed tree as a list."""
return self.parse().asList()
)
attr.leaveWhitespace()
attr.setName('attr')
hexdigits = Word(string.hexdigits, exact=2)
hexdigits.setName('hexdigits')
escaped = Suppress(Literal('\\')) + hexdigits
escaped.setName('escaped')
def _p_escaped(s, l, t):
text = t[0]
return chr(int(text, 16))
escaped.setParseAction(_p_escaped)
value = Combine(OneOrMore(CharsNotIn('*()\\\0') | escaped))
value.setName('value')
equal = Literal("=")
equal.setParseAction(lambda s, l, t: pureldap.LDAPFilter_equalityMatch)
approx = Literal("~=")
approx.setParseAction(lambda s, l, t: pureldap.LDAPFilter_approxMatch)
greater = Literal(">=")
greater.setParseAction(lambda s, l, t: pureldap.LDAPFilter_greaterOrEqual)
less = Literal("<=")
less.setParseAction(lambda s, l, t: pureldap.LDAPFilter_lessOrEqual)
filtertype = equal | approx | greater | less
filtertype.setName('filtertype')
simple = attr + filtertype + value
simple.leaveWhitespace()
simple.setName('simple')
class FilterExpression:
"""Filter factory based on filter expressions such as "name~John and (age>18 or consent:true)".
Filters consist of:
- boolean expressions using and, not, or and parentheses.
- field expressions, which consist of:
- a key name (with optional wildcards) or _index_
- an operator
- a value (with type appropriate to the operator)
- operators are one of:
- numeric comparisons: =, !=, <, <=, >, >=
- length comparisons: #=, #<, etc
- string comparisons: =, !=, ~ (regex match), !~
- containment: >> (contains), !>>
- existence: KEY:exists (has value that's not NaN or None), KEY:true (has true value)
"""
sign = oneOf("+ -")
integer = Word(nums)
number_base = (integer + Optional("." + Optional(integer))
("float")) | Literal(".")("float") + integer
number_exponent = CaselessLiteral("E")("float") + Optional(
sign) + integer
number = Combine(
Optional(sign) + number_base +
Optional(number_exponent)).setParseAction(
lambda t: float(t[0]) if t.float else int(t[0]))
onlen = lambda f: (lambda x, y: f(len(x), y))
num_ops = {
"<": operator.lt,
"<=": operator.le,
"=": operator.eq,
"!=": operator.ne,
">": operator.gt,
">=": operator.ge,
"#<": onlen(operator.lt),
"#<=": onlen(operator.le),
"#=": onlen(operator.eq),
"#!=": onlen(operator.ne),
"#>": onlen(operator.gt),
"#>=": onlen(operator.ge),
}
str_ops = {
"=": lambda x, y: x == str(y),
"!=": lambda x, y: x != str(y),
"~": lambda x, y: re.search(y, str(x)),
"!~": lambda x, y: not re.search(y, str(x)),
">>": lambda x, y: y in x,
"!>>": lambda x, y: y not in x,
}
exist_ops = {
":": lambda x, y: {
"exists": not non(x),
"true": bool(x)
}[y.lower()]
}
oneOfOpMap = lambda map: oneOf(list(map.keys())).setParseAction(
lambda t: ignoring_exceptions(map[t[0]], False))
num_op = oneOfOpMap(num_ops)
str_op = oneOfOpMap(str_ops)
exist_op = oneOfOpMap(exist_ops)
quoted_string = QuotedString('"', "\\") | QuotedString("'", "\\")
key_value = Word(alphas + alphas8bit + "*?[]_") | quoted_string
str_value = Word(alphas + alphas8bit + "_") | quoted_string
exist_value = CaselessLiteral("True") | CaselessLiteral("Exists")
base_expr = (key_value +
(str_op + str_value | num_op + number
| exist_op + exist_value)).setParseAction(lambda t: [t])
expr = infixNotation(
base_expr,
[
(Literal("not").setParseAction(lambda t: operator.not_), 1,
opAssoc.RIGHT),
(Literal("and").setParseAction(lambda t: operator.and_), 2,
opAssoc.LEFT),
(Literal("or").setParseAction(lambda t: operator.or_), 2,
opAssoc.LEFT),
],
)
@classmethod
def _eval_parse(cls, parse, d, i):
if not isinstance(parse, list):
return parse
elif len(parse) == 1:
return cls._eval_parse(parse[0], d, i)
elif callable(parse[0]):
return parse[0](cls._eval_parse(parse[1], d, i))
else:
x, y = cls._eval_parse(parse[0], d,
i), cls._eval_parse(parse[2], d, i)
if x == "_index_":
return parse[1](i, y)
elif isinstance(x, str):
return any(parse[1](d[k], y) for k in d.keys()
if fnmatch.fnmatch(k, x))
else:
return parse[1](x, y)
@classmethod
def make_filter(cls, string):
"""Generates a filter function from a filter expression."""
parse = cls.expr.parseString(string, parseAll=True).asList()
return lambda d, i=None: cls._eval_parse(parse, d, i)
def __init__(self):
from pyparsing import Word, nums, alphas, Combine, oneOf, opAssoc, operatorPrecedence
# Define the parser
integer = Word(nums).setParseAction(lambda t: int(t[0]))
real = Combine(Word(nums) + "." + Word(nums))
variable = Word(alphas, exact=1)
operand = real | integer | variable
# Operators
self.operators = {
'sign': oneOf('+ -'),
'multiply': oneOf('* /'),
'plus': oneOf('+ -'),
'comparision': oneOf('< <= > >= != = <> LT GT LE GE EQ NE'),
}
def operator_operands(token_list):
"""generator to extract operators and operands in pairs."""
it = iter(token_list)
while True:
try:
o1 = next(it)
o2 = next(it)
yield (o1, o2)
except StopIteration:
break
class EvalConstant(object):
"""Class to evaluate a parsed constant or variable."""
def __init__(self, tokens):
self.value = tokens[0]
def eval(self, vars):
if self.value in vars:
return vars[self.value]
else:
try:
return int(self.value)
except:
return float(self.value)
class EvalAddOp(object):
"""Class to evaluate addition and subtraction expressions."""
def __init__(self, tokens):
self.value = tokens[0]
def eval(self, vars):
sum = self.value[0].eval(vars)
for op, val in operator_operands(self.value[1:]):
if op == '+':
sum += val.eval(vars)
if op == '-':
sum -= val.eval(vars)
return sum
class EvalSignOp(object):
"""Class to evaluate expressions with a leading + or - sign."""
def __init__(self, tokens):
self.sign, self.value = tokens[0]
def eval(self, vars_):
mult = {'+': 1, '-': -1}[self.sign]
return mult * self.value.eval(vars_)
class EvalMultOp(object):
"""Class to evaluate multiplication and division expressions."""
def __init__(self, tokens):
self.operator_map = {
'*': lambda a, b: a * b,
'/': lambda a, b: a / b,
}
self.value = tokens[0]
def eval(self, vars):
prod = self.value[0].eval(vars)
for op, val in operator_operands(self.value[1:]):
fn = self.operator_map[op]
val2 = val.eval(vars)
prod = fn(prod, val2)
return prod
class EvalComparisonOp(object):
"""Class to evaluate comparison expressions"""
def __init__(self, tokens):
self.value = tokens[0]
self.operator_map = {
"<": lambda a, b: a < b,
"<=": lambda a, b: a <= b,
">": lambda a, b: a > b,
">=": lambda a, b: a >= b,
"!=": lambda a, b: a != b,
"=": lambda a, b: a == b,
"LT": lambda a, b: a < b,
"LE": lambda a, b: a <= b,
"GT": lambda a, b: a > b,
"GE": lambda a, b: a >= b,
"NE": lambda a, b: a != b,
"EQ": lambda a, b: a == b,
"<>": lambda a, b: a != b,
}
def eval(self, vars):
val1 = self.value[0].eval(vars)
for op, val in operator_operands(self.value[1:]):
fn = self.operator_map[op]
val2 = val.eval(vars)
if not fn(val1, val2):
break
val1 = val2
else:
return True
return False
operand.setParseAction(EvalConstant)
self.arith_expr = operatorPrecedence(operand, [
(self.operators['sign'], 1, opAssoc.RIGHT, EvalSignOp),
(self.operators['multiply'], 2, opAssoc.LEFT, EvalMultOp),
(self.operators['plus'], 2, opAssoc.LEFT, EvalAddOp),
(self.operators['comparision'], 2, opAssoc.LEFT, EvalComparisonOp),
])
# pyParsing tokens
# ---------------------------------------------------------------------
# Copyright (C) 2007-2015 The NOC Project
# See LICENSE for details
# ---------------------------------------------------------------------
# Third-party modules
from pyparsing import (alphanums, Combine, Group, LineEnd, nums, Suppress, Word,
restOfLine)
# Match \s+
SPACE = Suppress(Word(" ").leaveWhitespace())
# Match \n\s+
INDENT = Suppress(LineEnd() + SPACE)
# Skip whole line
LINE = Suppress(restOfLine)
#
REST = SPACE + restOfLine
# Sequence of numbers
DIGITS = Word(nums)
# Sequence of letters and numbers
ALPHANUMS = Word(alphanums)
# Number from 0 to 255
OCTET = Word(nums, max=3)
# IPv4 address
IPv4_ADDRESS = Combine(OCTET + "." + OCTET + "." + OCTET + "." + OCTET)
# RD
RD = Combine(Word(nums) + Word(":") + Word(nums))
if res:
res += [res[-3], rdflib.RDF.rest, b, b, rdflib.RDF.first, x]
else:
res += [b, rdflib.RDF.first, x]
res += [b, rdflib.RDF.rest, rdflib.RDF.nil]
if DEBUG:
print "CollectionOut", res
return [res]
# SPARQL Grammar from http://www.w3.org/TR/sparql11-query/#grammar
# ------ TERMINALS --------------
# [139] IRIREF ::= '<' ([^<>"{}|^`\]-[#x00-#x20])* '>'
IRIREF = Combine(
Suppress('<') + Regex(r'[^<>"{}|^`\\%s]*' % ''.join('\\x%02X' % i
for i in range(33))) +
Suppress('>'))
IRIREF.setParseAction(lambda x: rdflib.URIRef(x[0]))
# [164] P_CHARS_BASE ::= [A-Z] | [a-z] | [#x00C0-#x00D6] | [#x00D8-#x00F6] | [#x00F8-#x02FF] | [#x0370-#x037D] | [#x037F-#x1FFF] | [#x200C-#x200D] | [#x2070-#x218F] | [#x2C00-#x2FEF] | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD] | [#x10000-#xEFFFF]
if sys.maxunicode == 0xffff:
# this is narrow python build (default on windows/osx)
# this means that unicode code points over 0xffff are stored
# as several characters, which in turn means that regex character
# ranges with these characters do not work.
# See
# * http://bugs.python.org/issue12729
# * http://bugs.python.org/issue12749
# * http://bugs.python.org/issue3665
#
def __init__(self, funcname, selected_fields=False, userargs=None,
default_fields=None, parent=None, example=None, text=None):
"""funcname is name the function you want to use(can be either string, or functions.py function).
if combotags is true then a combobox with tags that the user can choose from are shown.
userargs is the default values you want to fill the controls in the dialog with
[make sure they don't exceed the number of arguments of funcname]."""
QWidget.__init__(self, parent)
identifier = QuotedString('"') | Combine(Word
(alphanums + ' !"#$%&\'()*+-./:;<=>?@[\\]^_`{|}~'))
tags = delimitedList(identifier)
self.func = Function(funcname)
docstr = self.func.doc[1:]
self.vbox = QVBoxLayout()
self.retval = []
self._selectedFields = selected_fields
if selected_fields:
fields = ['__all'] + sorted(INFOTAGS) + \
selected_fields + gettaglist()
else:
fields = ['__selected', '__all'] + sorted(INFOTAGS) + \
gettaglist()
self.tagcombo = QComboBox(self)
self.tagcombo.setToolTip(FIELDS_TOOLTIP)
self.tagcombo.setEditable(True)
self.tagcombo.setCompleter(QCompleter(self.tagcombo))
self.tagcombo.addItems(fields)
self.tagcombo.editTextChanged.connect(self.showexample)
if self.func.function not in functions.no_fields:
label = QLabel(translate('Defaults', "&Fields"))
self.vbox.addWidget(label)
self.vbox.addWidget(self.tagcombo)
label.setBuddy(self.tagcombo)
else:
self.tagcombo.setVisible(False)
self.example = example
self._text = text
if self.func.function in functions_dialogs.dialogs:
vbox = QVBoxLayout()
vbox.addWidget(self.tagcombo)
self.widget = functions_dialogs.dialogs[self.func.function](self)
vbox.addWidget(self.widget)
vbox.addStretch()
self.setLayout(vbox)
self.setMinimumSize(self.sizeHint())
self.setArguments(default_fields, userargs)
return
else:
self.widget = None
self.textcombos = []
# Loop that creates all the controls
self.controls = []
for argno, line in enumerate(docstr):
args = tags.parseString(line)
label = args[0]
ctype = args[1]
default = args[2:]
control, func, label = self._createControl(label, ctype, default)
self.retval.append(func)
self.controls.append(control)
getattr(control, self.signals[ctype]).connect(self.showexample)
if label:
self.vbox.addWidget(label)
self.vbox.addWidget(control)
self.setArguments(default_fields, userargs)
self.vbox.addStretch()
self.setLayout(self.vbox)
self.setMinimumSize(self.sizeHint())
# The aim of this parser is not to support database application,
# but to create automagically a pgn annotated reading the log console file
# of a lecture of ICC (Internet Chess Club), saved by Blitzin.
# Of course you can modify the Abstract Syntax Tree to your purpose.
#
# Copyright 2004, by Alberto Santini http://www.albertosantini.it/chess/
#
from pyparsing import alphanums, nums, quotedString
from pyparsing import Combine, Forward, Group, Literal, oneOf, OneOrMore, Optional, Suppress, ZeroOrMore, White, Word
from pyparsing import ParseException
#
# define pgn grammar
#
tag = Suppress("[") + Word(alphanums) + Combine(quotedString) + Suppress("]")
comment = Suppress("{") + Word(alphanums + " ") + Suppress("}")
dot = Literal(".")
piece = oneOf("K Q B N R")
file_coord = oneOf("a b c d e f g h")
rank_coord = oneOf("1 2 3 4 5 6 7 8")
capture = oneOf("x :")
promote = Literal("=")
castle_queenside = oneOf("O-O-O 0-0-0 o-o-o")
castle_kingside = oneOf("O-O 0-0 o-o")
move_number = Optional(comment) + Word(nums) + dot
m1 = file_coord + rank_coord # pawn move e.g. d4
m2 = file_coord + capture + file_coord + rank_coord # pawn capture move e.g. dxe5
m3 = file_coord + "8" + promote + piece # pawn promotion e.g. e8=Q
# Grammar Rules
# ============================================================================ #
mul = Literal("*")
plus = Literal("+")
minus = Literal("-")
comma = Literal(",")
lbracket = Literal("[")
rbracket = Literal("]")
lbrace = Literal("{")
rbrace = Literal("}")
hashsign = Literal("#")
exclamation = Literal("!")
caret = Literal("^")
hex_num = Combine(Literal("0x") + Word("0123456789abcdef"))
dec_num = Word("0123456789")
# Operand Parsing
# ============================================================================ #
sign = Optional(Or([plus, minus("minus")]))
immediate = Group(
Optional(Suppress(hashsign)) + (sign + Or([hex_num, dec_num]))("value"))
register = Group(
Or([
Combine(Literal("r") + Word(nums)("reg_num")),
Combine(Literal("d") + Word(nums)("reg_num")),
Combine(Literal("c") + Word(nums)("reg_num")),
Combine(Literal("p") + Word(nums)("reg_num")),
except ImportError:
ecodes = None
print('WARNING: evdev is not available')
try:
from functools import lru_cache
except ImportError:
# don't do caching on old python
lru_cache = lambda: (lambda f: f)
EOL = LineEnd().suppress()
EMPTYLINE = LineEnd()
COMMENTLINE = pythonStyleComment + EOL
INTEGER = Word(nums)
STRING = QuotedString('"')
REAL = Combine((INTEGER + Optional('.' + Optional(INTEGER))) ^ ('.' + INTEGER))
SIGNED_REAL = Combine(Optional(Word('-+')) + REAL)
UDEV_TAG = Word(string.ascii_uppercase, alphanums + '_')
# Those patterns are used in type-specific matches
TYPES = {'mouse': ('usb', 'bluetooth', 'ps2', '*'),
'evdev': ('name', 'atkbd', 'input'),
'id-input': ('modalias'),
'touchpad': ('i8042', 'rmi', 'bluetooth', 'usb'),
'joystick': ('i8042', 'rmi', 'bluetooth', 'usb'),
'keyboard': ('name', ),
'sensor': ('modalias', ),
}
# Patterns that are used to set general properties on a device
GENERAL_MATCHES = {'acpi',
def create_bnf(term_descs):
"""term_descs .. list of TermParse objects
(sign, term_name, term_arg_names), where sign can be real or complex
multiplier"""
lc = ['+'] # Linear combination context.
equal = Literal("=").setParseAction(rhs(lc))
zero = Literal("0").suppress()
point = Literal(".")
e = CaselessLiteral("E")
inumber = Word("+-" + nums, nums)
fnumber = Combine(
Word("+-" + nums, nums) + Optional(point + Optional(Word(nums))) +
Optional(e + Word("+-" + nums, nums)))
number = fnumber + Optional(Literal('j'), default='')
add_op = oneOf('+ -')
number_expr = Forward()
number_expr << ZeroOrMore('(') + number \
+ ZeroOrMore(add_op + number_expr) \
+ ZeroOrMore(')')
ident = Word(alphas, alphanums + "_")
integral = Combine((Literal('i') + Word(alphanums)) | Literal('a')
| Word(nums))("integral")
history = Optional('[' + inumber + ']', default='')("history")
variable = Combine(Word(alphas, alphanums + '._') + history)
derivative = Combine(Literal('d') + variable\
+ Literal('/') + Literal('dt'))
trace = Combine(Literal('tr') + '(' + variable + ')')
generalized_var = derivative | trace | variable
args = Group(delimitedList(generalized_var))
flag = Literal('a')
term = Optional( Literal( '+' ) | Literal( '-' ), default = '+' )( "sign" )\
+ Optional( number_expr + Literal( '*' ).suppress(),
default = ['1.0', ''] )( "mul" ) \
+ Combine( ident( "name" )\
+ Optional( "." + (integral + "."
+ ident( "region" ) + "." + flag( "flag" ) |
integral + "." + ident( "region" ) |
ident( "region" )
)))( "term_desc" ) + "("\
+ Optional(args, default=[''])( "args" ) + ")"
term.setParseAction(collect_term(term_descs, lc))
rhs1 = equal + OneOrMore(term)
rhs2 = equal + zero
equation = StringStart() + OneOrMore( term )\
+ Optional( rhs1 | rhs2 ) + StringEnd()
## term.setDebug()
return equation
escaped = (
Literal("\\").suppress()
+
# chr(20)-chr(126) + chr(128)-unichr(sys.maxunicode)
Regex("[\u0020-\u007e\u0080-\uffff]", re.IGNORECASE)
)
def convertToUnicode(t):
return chr(int(t[0], 16))
hex_unicode = (
Literal("\\").suppress()
+ Regex("[0-9a-f]{1,6}", re.IGNORECASE)
+ Optional(White(exact=1)).suppress()
).setParseAction(convertToUnicode)
escape = hex_unicode | escaped
# any unicode literal outside the 0-127 ascii range
nonascii = Regex("[^\u0000-\u007f]")
# single character for starting an identifier.
nmstart = Regex("[A-Z]", re.IGNORECASE) | nonascii | escape
nmchar = Regex("[0-9A-Z-]", re.IGNORECASE) | nonascii | escape
identifier = Combine(nmstart + ZeroOrMore(nmchar))
binopstr = " ".join(opcodes)
def parsebinop(opexpr):
"parse action for binary operators"
left, opstr, right = opexpr
for opclass in binopclasses:
if opstr in opclass.literals:
return opclass(left, right)
binop = oneOf(binopstr)
arithSign = Word("+-", exact=1)
realNum = Combine(Optional(arithSign) +
(Word(nums) + "." + Optional(Word(nums)) | ("." + Word(nums))) # noqa
+ Optional(E + Optional(arithSign) + Word(nums)))
realNum.setParseAction(lambda x: expression.NumericLiteral(float(x[0])))
intNum = Combine(Optional(arithSign) + Word(nums) +
Optional(E + Optional("+") + Word(nums)))
intNum.setParseAction(lambda x: expression.NumericLiteral(int(x[0])))
number = realNum | intNum
variable = ident.copy()
variable.setParseAction(lambda x: model.Var(x[0]))
quotedString.setParseAction(lambda x: expression.StringLiteral(x[0][1:-1]))
literal = quotedString | number
from pyparsing import OneOrMore
from pyparsing import Word
from pyparsing import ZeroOrMore
from pyparsing import alphanums
from pyparsing import nums
# Intermediate parsers
_varchar_names = (CaselessKeyword('VARCHAR') | CaselessKeyword('TEXT'))
_varchar_names |= CaselessKeyword('NVARCHAR')
# Data types
_smallint = (CaselessKeyword('SMALLINT') | CaselessKeyword('INT2'))
_integer = CaselessKeyword('INTEGER')
_integer |= CaselessKeyword('INT') | CaselessKeyword('INT4')
_bigint = (CaselessKeyword('BIGINT') | CaselessKeyword('INT8'))
_decimal = Combine((CaselessKeyword('DECIMAL') | CaselessKeyword('NUMERIC')) + '(' + Word(nums + ' ,') + ')') # noqa
_real = (CaselessKeyword('REAL') | CaselessKeyword('FLOAT4'))
_double = (CaselessKeyword('DOUBLE PRECISION') | CaselessKeyword('FLOAT') | CaselessKeyword('FLOAT8') | CaselessKeyword('DOUBLE')) # noqa
_boolean = CaselessKeyword('BOOLEAN')
_char = (CaselessKeyword('CHAR') | CaselessKeyword('CHARACTER'))
_char |= (CaselessKeyword('NCHAR') | CaselessKeyword('BPCHAR'))
_varchar = Combine(_varchar_names + '(' + Word(alphanums) + ')')
_date = CaselessKeyword('DATE')
_text = CaselessKeyword('TEXT')
_timestamp = CaselessKeyword('TIMESTAMP')
# Create SQL keywords
_create = CaselessKeyword('CREATE')
_table = CaselessKeyword('TABLE')
_view = CaselessKeyword('VIEW')
_temp = CaselessKeyword('TEMP')
def __init__(self):
self.ae = False
self.local_dict = None
self.f = None
self.user_functions = None
self.expr_stack = []
self.texpr_stack = []
# Define constants
self.constants = {}
# Define Operators
self.opn = {"+": operator.add,
"-": operator.sub,
"*": operator.mul,
"/": operator.truediv,
">": operator.gt,
">=": operator.ge,
"<": operator.lt,
"<=": operator.le,
"==": operator.eq,
"!=": operator.ne,
"|": operator.or_,
"&": operator.and_,
"!": operator.inv}
# Define xarray DataArray operators with 1 input parameter
self.xfn1 = {"angle": xr.ufuncs.angle,
"arccos": xr.ufuncs.arccos,
"arccosh": xr.ufuncs.arccosh,
"arcsin": xr.ufuncs.arcsin,
"arcsinh": xr.ufuncs.arcsinh,
"arctan": xr.ufuncs.arctan,
"arctanh": xr.ufuncs.arctanh,
"ceil": xr.ufuncs.ceil,
"conj": xr.ufuncs.conj,
"cos": xr.ufuncs.cos,
"cosh": xr.ufuncs.cosh,
"deg2rad": xr.ufuncs.deg2rad,
"degrees": xr.ufuncs.degrees,
"exp": xr.ufuncs.exp,
"expm1": xr.ufuncs.expm1,
"fabs": xr.ufuncs.fabs,
"fix": xr.ufuncs.fix,
"floor": xr.ufuncs.floor,
"frexp": xr.ufuncs.frexp,
"imag": xr.ufuncs.imag,
"iscomplex": xr.ufuncs.iscomplex,
"isfinite": xr.ufuncs.isfinite,
"isinf": xr.ufuncs.isinf,
"isnan": xr.ufuncs.isnan,
"isreal": xr.ufuncs.isreal,
"log": xr.ufuncs.log,
"log10": xr.ufuncs.log10,
"log1p": xr.ufuncs.log1p,
"log2": xr.ufuncs.log2,
"rad2deg": xr.ufuncs.rad2deg,
"radians": xr.ufuncs.radians,
"real": xr.ufuncs.real,
"rint": xr.ufuncs.rint,
"sign": xr.ufuncs.sign,
"signbit": xr.ufuncs.signbit,
"sin": xr.ufuncs.sin,
"sinh": xr.ufuncs.sinh,
"sqrt": xr.ufuncs.sqrt,
"square": xr.ufuncs.square,
"tan": xr.ufuncs.tan,
"tanh": xr.ufuncs.tanh,
"trunc": xr.ufuncs.trunc}
# Define xarray DataArray operators with 2 input parameter
self.xfn2 = {"arctan2": xr.ufuncs.arctan2,
"copysign": xr.ufuncs.copysign,
"fmax": xr.ufuncs.fmax,
"fmin": xr.ufuncs.fmin,
"fmod": xr.ufuncs.fmod,
"hypot": xr.ufuncs.hypot,
"ldexp": xr.ufuncs.ldexp,
"logaddexp": xr.ufuncs.logaddexp,
"logaddexp2": xr.ufuncs.logaddexp2,
"logicaland": xr.ufuncs.logical_and,
"logicalnot": xr.ufuncs.logical_not,
"logicalor": xr.ufuncs.logical_or,
"logicalxor": xr.ufuncs.logical_xor,
"maximum": xr.ufuncs.maximum,
"minimum": xr.ufuncs.minimum,
"nextafter": xr.ufuncs.nextafter}
# Define non-xarray DataArray operators with 2 input parameter
self.fn2 = {"percentile": np.percentile}
# Define xarray DataArray reduction operators
self.xrfn = {"all": xr.DataArray.all,
"any": xr.DataArray.any,
"argmax": xr.DataArray.argmax,
"argmin": xr.DataArray.argmin,
"max": xr.DataArray.max,
"mean": xr.DataArray.mean,
"median": xr.DataArray.median,
"min": xr.DataArray.min,
"prod": xr.DataArray.prod,
"sum": xr.DataArray.sum,
"std": xr.DataArray.std,
"var": xr.DataArray.var}
# Define non-xarray DataArray operators with 2 input parameter
self.xcond = {"<": np.percentile}
# Define Grammar
point = Literal(".")
e = CaselessLiteral("E")
fnumber = Combine(Word("+-"+nums, nums) +
Optional(point + Optional(Word(nums))) +
Optional(e + Word("+-"+nums, nums)))
variable = Word(alphas, alphas+nums+"_$")
seq = Literal("=")
b_not = Literal("~")
plus = Literal("+")
minus = Literal("-")
mult = Literal("*")
div = Literal("/")
gt = Literal(">")
gte = Literal(">=")
lt = Literal("<")
lte = Literal("<=")
eq = Literal("==")
neq = Literal("!=")
b_or = Literal("|")
b_and = Literal("&")
l_not = Literal("!")
lpar = Literal("(").suppress()
rpar = Literal(")").suppress()
comma = Literal(",")
colon = Literal(":")
lbrac = Literal("[")
rbrac = Literal("]")
lcurl = Literal("{")
rcurl = Literal("}")
qmark = Literal("?")
scolon = Literal(";")
addop = plus | minus
multop = mult | div
sliceop = colon
compop = gte | lte | gt | lt
eqop = eq | neq
bitcompop = b_or | b_and
bitnotop = b_not
logicalnotop = l_not
assignop = seq
expop = Literal("^")
expr = Forward()
indexexpr = Forward()
atom = (Optional("-") +
(variable + seq + expr).setParseAction(self.push_assign) |
indexexpr.setParseAction(self.push_index) |
(lpar + expr + qmark.setParseAction(self.push_ternary1) + expr +
scolon.setParseAction(self.push_ternary2) + expr +
rpar).setParseAction(self.push_ternary) |
(lpar + expr + qmark + expr + scolon + expr +
rpar).setParseAction(self.push_ternary) |
(logicalnotop + expr).setParseAction(self.push_ulnot) |
(bitnotop + expr).setParseAction(self.push_unot) |
(minus + expr).setParseAction(self.push_uminus) |
(variable + lcurl + expr +
rcurl).setParseAction(self.push_mask) |
(variable + lpar + expr + (comma + expr)*3 +
rpar).setParseAction(self.push_expr4) |
(variable + lpar + expr + (comma + expr)*2 +
rpar).setParseAction(self.push_expr3) |
(variable + lpar + expr + comma + expr +
rpar).setParseAction(self.push_expr2) |
(variable + lpar + expr + rpar |
variable).setParseAction(self.push_expr1) |
fnumber.setParseAction(self.push_expr) |
(lpar + expr + ZeroOrMore(comma + expr).setParseAction(self.get_tuple) +
rpar).setParseAction(self.push_tuple) |
(lpar + expr.suppress() +
rpar).setParseAction(self.push_uminus))
# Define order of operations for operators
factor = Forward()
factor << atom + ZeroOrMore((expop + factor).setParseAction(self.push_op))
term = factor + ZeroOrMore((multop + factor).setParseAction(self.push_op))
term2 = term + ZeroOrMore((addop + term).setParseAction(self.push_op))
term3 = term2 + ZeroOrMore((sliceop + term2).setParseAction(self.push_op))
term4 = term3 + ZeroOrMore((compop + term3).setParseAction(self.push_op))
term5 = term4 + ZeroOrMore((eqop + term4).setParseAction(self.push_op))
term6 = term5 + ZeroOrMore((bitcompop + term5).setParseAction(self.push_op))
expr << term6 + ZeroOrMore((assignop + term6).setParseAction(self.push_op))
# Define index operators
colon_expr = (colon + FollowedBy(comma) ^ colon +
FollowedBy(rbrac)).setParseAction(self.push_colon)
range_expr = colon_expr | expr | colon
indexexpr << (variable + lbrac + delimitedList(range_expr, delim=',') +
rbrac).setParseAction(self.push_expr)
self.parser = expr
dot_fanging_patterns = Combine(
Optional(White()) + Or([
# '.' - enclosed with ( and )
CaselessLiteral("(.("),
CaselessLiteral("(.)"),
CaselessLiteral(").("),
CaselessLiteral(").)"),
CaselessLiteral("(."),
CaselessLiteral(".("),
# CaselessLiteral(")."), # this is commented and is NOT used to fang indicators b/c this may appear in real text
CaselessLiteral(".)"),
# 'dot' - enclosed with ( and )
CaselessLiteral("(dot("),
CaselessLiteral("(dot)"),
CaselessLiteral(")dot("),
CaselessLiteral(")dot)"),
CaselessLiteral("(dot"),
CaselessLiteral("dot("),
CaselessLiteral(")dot"),
CaselessLiteral("dot)"),
# 'punkt' - enclosed with ( and )
CaselessLiteral("(punkt("),
CaselessLiteral("(punkt)"),
CaselessLiteral(")punkt("),
CaselessLiteral(")punkt)"),
CaselessLiteral("(punkt"),
CaselessLiteral("punkt("),
CaselessLiteral(")punkt"),
CaselessLiteral("punkt)"),
# 'punto' - enclosed with ( and )
CaselessLiteral("(punto("),
CaselessLiteral("(punto)"),
CaselessLiteral(")punto("),
CaselessLiteral(")punto)"),
CaselessLiteral("(punto"),
CaselessLiteral("punto("),
CaselessLiteral(")punto"),
CaselessLiteral("punto)"),
# '.' - enclosed with [ and ]
CaselessLiteral("[.["),
CaselessLiteral("[.]"),
CaselessLiteral("].["),
CaselessLiteral("].]"),
CaselessLiteral("[."),
CaselessLiteral(".["),
CaselessLiteral("]."),
CaselessLiteral(".]"),
# 'dot' - enclosed with [ and ]
CaselessLiteral("[dot["),
CaselessLiteral("[dot]"),
CaselessLiteral("]dot["),
CaselessLiteral("]dot]"),
CaselessLiteral("[dot"),
CaselessLiteral("dot["),
CaselessLiteral("]dot"),
CaselessLiteral("dot]"),
# 'punkt' - enclosed with [ and ]
CaselessLiteral("[punkt["),
CaselessLiteral("[punkt]"),
CaselessLiteral("]punkt["),
CaselessLiteral("]punkt]"),
CaselessLiteral("[punkt"),
CaselessLiteral("punkt["),
CaselessLiteral("]punkt"),
CaselessLiteral("punkt]"),
# 'punto' - enclosed with [ and ]
CaselessLiteral("[punto["),
CaselessLiteral("[punto]"),
CaselessLiteral("]punto["),
CaselessLiteral("]punto]"),
CaselessLiteral("[punto"),
CaselessLiteral("punto["),
CaselessLiteral("]punto"),
CaselessLiteral("punto]"),
# '.' - enclosed with { and }
CaselessLiteral("{.{"),
CaselessLiteral("{.}"),
CaselessLiteral("}.{"),
CaselessLiteral("}.}"),
CaselessLiteral("{."),
CaselessLiteral(".{"),
CaselessLiteral("}."),
CaselessLiteral(".}"),
# 'dot' - enclosed with { and }
CaselessLiteral("{dot{"),
CaselessLiteral("{dot}"),
CaselessLiteral("}dot{"),
CaselessLiteral("}dot}"),
CaselessLiteral("{dot"),
CaselessLiteral("dot{"),
CaselessLiteral("}dot"),
CaselessLiteral("dot}"),
# 'punkt' - enclosed with { and }
CaselessLiteral("{punkt{"),
CaselessLiteral("{punkt}"),
CaselessLiteral("}punkt{"),
CaselessLiteral("}punkt}"),
CaselessLiteral("{punkt"),
CaselessLiteral("punkt{"),
CaselessLiteral("}punkt"),
CaselessLiteral("punkt}"),
# 'punto' - enclosed with { and }
CaselessLiteral("{punto{"),
CaselessLiteral("{punto}"),
CaselessLiteral("}punto{"),
CaselessLiteral("}punto}"),
CaselessLiteral("{punto"),
CaselessLiteral("punto{"),
CaselessLiteral("}punto"),
CaselessLiteral("punto}"),
# a
Literal("DOT"),
Literal("PUNKT"),
Literal("PUNTO"),
CaselessLiteral("-dot-"),
CaselessLiteral("-punkt-"),
CaselessLiteral("-punto-"),
]) + Optional(White())).addParseAction(replaceWith("."))
aop0 = oneOf('* /')
aop1 = oneOf('+ -')
aop2 = oneOf('%').setParseAction(lambda s, l, t: ['mod'])
bop = oneOf('& |').setParseAction(lambda s, l, t: ['and']
if t[0] == '&' else ['or'])
NOT = Literal('!')
rop = oneOf('< > <= >= = !=').setParseAction(lambda s, l, t: ['distinct']
if t[0] == '!=' else t)
GET, CAT, HAS, IND, LEN, REP, SUB, EQL = map(
Literal, '#get #cat #has #ind #len #rep #sub #eql'.split())
var = Word(alphas + '_:$', alphanums + '_:$').setParseAction(addVar)
ival = Combine(Optional('-') + Word(nums)).setParseAction(
lambda s, l, t: ['(- %s)' % t[0][1:]] if t[0][0] == '-' else t)
ivar = (ival + var).setParseAction(lambda s, l, t: ['*', t[0], t[1]])
term = ivar | ival | var | QuotedString(quoteChar='"', unquoteResults=False)
stmt = Forward()
expr = Forward()
sexpr = Forward()
sexpr << (
(GET + LPAR + expr + COMMA + expr +
RPAR).setParseAction(lambda s, l, t: CharAtAction(t))
| (CAT + LPAR + expr + COMMA + expr + RPAR).setParseAction(
lambda s, l, t: [['Concat', chkString(t[1]),
chkString(t[2])]])
|
def define_dot_parser(self):
"""Define dot grammar
Based on the grammar http://www.graphviz.org/doc/info/lang.html
"""
# punctuation
colon = Literal(":")
lbrace = Suppress("{")
rbrace = Suppress("}")
lbrack = Suppress("[")
rbrack = Suppress("]")
lparen = Literal("(")
rparen = Literal(")")
equals = Suppress("=")
comma = Literal(",")
dot = Literal(".")
slash = Literal("/")
bslash = Literal("\\")
star = Literal("*")
semi = Suppress(";")
at = Literal("@")
minus = Literal("-")
pluss = Suppress("+")
# keywords
strict_ = CaselessLiteral("strict")
graph_ = CaselessLiteral("graph")
digraph_ = CaselessLiteral("digraph")
subgraph_ = CaselessLiteral("subgraph")
node_ = CaselessLiteral("node")
edge_ = CaselessLiteral("edge")
punctuation_ = "".join([c for c in string.punctuation if c not in '_'
]) + string.whitespace
# token definitions
identifier = Word(alphanums + "_").setName("identifier")
#double_quoted_string = QuotedString('"', multiline=True,escChar='\\',
# unquoteResults=True) # dblQuotedString
double_quoted_string = Regex(r'\"(?:\\\"|\\\\|[^"])*\"', re.MULTILINE)
double_quoted_string.setParseAction(removeQuotes)
quoted_string = Combine(
double_quoted_string +
Optional(OneOrMore(pluss + double_quoted_string)),
adjacent=False)
alphastring_ = OneOrMore(CharsNotIn(punctuation_))
def parse_html(s, loc, toks):
return '<<%s>>' % ''.join(toks[0])
opener = '<'
closer = '>'
try:
html_text = pyparsing.nestedExpr(
opener, closer,
((CharsNotIn(opener + closer).setParseAction(lambda t: t[0]))
)).setParseAction(parse_html)
except:
log.debug('nestedExpr not available.')
log.warning('Old version of pyparsing detected. Version 1.4.8 or '
'later is recommended. Parsing of html labels may not '
'work properly.')
html_text = Combine(Literal("<<") + OneOrMore(CharsNotIn(",]")))
ID = (
alphastring_ | html_text | quoted_string
| #.setParseAction(strip_quotes) |
identifier).setName("ID")
float_number = Combine(Optional(minus) +
OneOrMore(Word(nums +
"."))).setName("float_number")
righthand_id = (float_number | ID).setName("righthand_id")
port_angle = (at + ID).setName("port_angle")
port_location = ((OneOrMore(Group(colon + ID))
| Group(colon + lparen + ID + comma + ID +
rparen))).setName("port_location")
port = Combine(
(Group(port_location + Optional(port_angle))
| Group(port_angle + Optional(port_location)))).setName("port")
node_id = (ID + Optional(port))
a_list = OneOrMore(ID + Optional(equals + righthand_id) +
Optional(comma.suppress())).setName("a_list")
attr_list = OneOrMore(lbrack + Optional(a_list) + rbrack).setName(
"attr_list").setResultsName('attrlist')
attr_stmt = ((graph_ | node_ | edge_) + attr_list).setName("attr_stmt")
edgeop = (Literal("--") | Literal("->")).setName("edgeop")
stmt_list = Forward()
graph_stmt = (lbrace + Optional(stmt_list) + rbrace +
Optional(semi)).setName("graph_stmt")
edge_point = Forward()
edgeRHS = OneOrMore(edgeop + edge_point)
edge_stmt = edge_point + edgeRHS + Optional(attr_list)
subgraph = (
Optional(subgraph_, '') + Optional(ID, '') +
Group(graph_stmt)).setName("subgraph").setResultsName('ssubgraph')
edge_point << (subgraph | graph_stmt | node_id)
node_stmt = (node_id + Optional(attr_list) +
Optional(semi)).setName("node_stmt")
assignment = (ID + equals + righthand_id).setName("assignment")
stmt = (assignment | edge_stmt | attr_stmt | subgraph | graph_stmt
| node_stmt).setName("stmt")
stmt_list << OneOrMore(stmt + Optional(semi))
graphparser = ((Optional(strict_, 'notstrict') +
((graph_ | digraph_)) + Optional(ID, '') + lbrace +
Group(Optional(stmt_list)) +
rbrace).setResultsName("graph"))
singleLineComment = Group("//" + restOfLine) | Group("#" + restOfLine)
# actions
graphparser.ignore(singleLineComment)
graphparser.ignore(cStyleComment)
node_id.setParseAction(self._proc_node_id)
assignment.setParseAction(self._proc_attr_assignment)
a_list.setParseAction(self._proc_attr_list)
edge_stmt.setParseAction(self._proc_edge_stmt)
node_stmt.setParseAction(self._proc_node_stmt)
attr_stmt.setParseAction(self._proc_default_attr_stmt)
attr_list.setParseAction(self._proc_attr_list_combine)
subgraph.setParseAction(self._proc_subgraph_stmt)
#graph_stmt.setParseAction(self._proc_graph_stmt)
graphparser.setParseAction(self._main_graph_stmt)
return graphparser
b = rdflib.BNode()
if res:
res += [res[-3], rdflib.RDF.rest, b, b, rdflib.RDF.first, x]
else:
res += [b, rdflib.RDF.first, x]
res += [b, rdflib.RDF.rest, rdflib.RDF.nil]
if DEBUG:
print "CollectionOut", res
return [res]
# SPARQL Grammar from http://www.w3.org/TR/sparql11-query/#grammar
# ------ TERMINALS --------------
# [139] IRIREF ::= '<' ([^<>"{}|^`\]-[#x00-#x20])* '>'
IRIREF = Combine(Suppress("<") + Regex(r'[^<>"{}|^`\\%s]*' % "".join("\\x%02X" % i for i in range(33))) + Suppress(">"))
IRIREF.setParseAction(lambda x: rdflib.URIRef(x[0]))
# [164] P_CHARS_BASE ::= [A-Z] | [a-z] | [#x00C0-#x00D6] | [#x00D8-#x00F6] | [#x00F8-#x02FF] | [#x0370-#x037D] | [#x037F-#x1FFF] | [#x200C-#x200D] | [#x2070-#x218F] | [#x2C00-#x2FEF] | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD] | [#x10000-#xEFFFF]
if sys.maxunicode == 0xFFFF:
# this is narrow python build (default on windows/osx)
# this means that unicode code points over 0xffff are stored
# as several characters, which in turn means that regex character
# ranges with these characters do not work.
# See
# * http://bugs.python.org/issue12729
# * http://bugs.python.org/issue12749
# * http://bugs.python.org/issue3665
#
# Here we simple skip the [#x10000-#xEFFFF] part
def __init__(self):
"""
expop :: '^'
multop :: 'x' | '/'
addop :: '+' | '-'
integer :: ['+' | '-'] '0'..'9'+
atom :: PI | E | real | fn '(' expr ')' | '(' expr ')'
factor :: atom [ expop factor ]*
term :: factor [ multop factor ]*
expr :: term [ addop term ]*
"""
point = Literal(".")
exp = CaselessLiteral("E")
fnumber = Combine(
Word("+-" + nums, nums) + Optional(point + Optional(Word(nums))) +
Optional(exp + Word("+-" + nums, nums)))
ident = Word(alphas, alphas + nums + "_$")
plus = Literal("+")
minus = Literal("-")
mult = Literal("x")
div = Literal("/")
lpar = Literal("(").suppress()
rpar = Literal(")").suppress()
addop = plus | minus
multop = mult | div
powop = Literal("^")
pi = CaselessLiteral("PI")
expr = Forward()
atom = (
(Optional(oneOf("- +")) +
(pi | exp | fnumber | ident + lpar + expr + rpar).setParseAction(
self.push_first))
| Optional(oneOf("- +")) +
Group(lpar + expr + rpar)).setParseAction(self.push_unary_minus)
# by defining exponentiation as "atom [ ^ factor ]..." instead of
# "atom [ ^ atom ]...", we get right-to-left exponents, instead of left-to-right
# that is, 2^3^2 = 2^(3^2), not (2^3)^2.
factor = Forward()
factor << atom + ZeroOrMore(
(powop + factor).setParseAction(self.push_first))
term = factor + ZeroOrMore(
(multop + factor).setParseAction(self.push_first))
expr << term + ZeroOrMore(
(addop + term).setParseAction(self.push_first))
# addop_term = ( addop + term ).setParseAction( self.pushFirst )
# general_term = term + ZeroOrMore( addop_term ) | OneOrMore( addop_term)
# expr << general_term
self.bnf = expr
# map operator symbols to corresponding arithmetic operations
epsilon = 1e-12
self.opn = {
"+": operator.add,
"-": operator.sub,
"x": operator.mul,
"/": operator.truediv,
"^": operator.pow
}
self.function = {
"sin": math.sin,
"cos": math.cos,
"tan": math.tan,
"abs": abs,
"trunc": lambda a: int(a),
"round": round,
"sgn": lambda a: abs(a) > epsilon and cmp(a, 0) or 0
}
def _grammar():
from pyparsing import alphas, alphanums, nums
from pyparsing import oneOf, Suppress, Optional, Group, ZeroOrMore, NotAny
from pyparsing import Forward, operatorPrecedence, opAssoc, Word, White
from pyparsing import delimitedList, Combine, Literal, OneOrMore
expression = Forward()
LPAR, RPAR, DOT, LBRAC, RBRAC = map(Suppress, "().{}")
nw = NotAny(White())
identifier = Word(alphas + "_", alphanums + "_")
integer = Word(nums)
integer.setParseAction(IntegerNode)
fractional = Combine(Word('+' + '-' + nums, nums) + '.' + Word(nums))
fractional.setParseAction(FloatNode)
literal = fractional | integer
arglist = delimitedList(expression)
seqrange = LBRAC + expression + Suppress('..') + expression + RBRAC
seqrange.setParseAction(lambda t: SequenceNode(start=t[0], stop=t[1]))
seqexplicit = LBRAC + Optional(arglist) + RBRAC
seqexplicit.setParseAction(lambda t: SequenceNode(lst=t))
sequence = seqrange | seqexplicit
rollmod = nw + Group(oneOf("d k r e x") + Optional(integer))
numdice = Optional(integer, default=1)
roll = numdice + nw + Suppress("d") + nw + (integer | sequence)
roll += Group(ZeroOrMore(rollmod))
roll.setParseAction(DieRollNode)
call = LPAR + Group(Optional(arglist)) + RPAR
function = identifier + call
function.setParseAction(FunctionNode)
seqexpr = ((roll | sequence | function)
+ Group(OneOrMore(DOT + identifier + call)))
seqexpr.setParseAction(SeqMethodNode)
variable = Word(alphas + "_", alphanums + "_ ")
variable.setParseAction(VariableNode)
atom = seqexpr | roll | literal | sequence | function | variable
expoop = Literal('^')
signop = oneOf("+ -")
multop = oneOf("* /")
plusop = oneOf("+ -")
# noinspection PyUnresolvedReferences
expression << operatorPrecedence(
atom,
[
(expoop, 2, opAssoc.LEFT, BinaryOpNode),
(signop, 1, opAssoc.RIGHT, UnaryOpNode),
(multop, 2, opAssoc.LEFT, BinaryOpNode),
(plusop, 2, opAssoc.LEFT, BinaryOpNode),
]
)
return expression
def rc_statement():
"""
Generate a RC statement parser that can be used to parse a RC file
:rtype: pyparsing.ParserElement
"""
one_line_comment = '//' + restOfLine
comments = cStyleComment ^ one_line_comment
precompiler = Word('#', alphanums) + restOfLine
language_definition = "LANGUAGE" + Word(alphas + '_').setResultsName(
"language") + Optional(',' + Word(alphas +
'_').setResultsName("sublanguage"))
block_start = (Keyword('{') | Keyword("BEGIN")).setName("block_start")
block_end = (Keyword('}') | Keyword("END")).setName("block_end")
reserved_words = block_start | block_end
name_id = ~reserved_words + \
Word(alphas, alphanums + '_').setName("name_id")
numbers = Word(nums)
integerconstant = numbers ^ Combine('0x' + numbers)
constant = Combine(Optional(Keyword("NOT")) + (name_id | integerconstant),
adjacent=False,
joinString=' ')
combined_constants = delimitedList(constant, '|')
block_options = Optional(
SkipTo(Keyword("CAPTION"), failOn=block_start)("pre_caption") +
Keyword("CAPTION") +
quotedString("caption")) + SkipTo(block_start)("post_caption")
undefined_control = Group(
name_id.setResultsName("id_control") +
delimitedList(quotedString ^ constant ^ numbers
^ Group(combined_constants)).setResultsName("values_"))
block = block_start + \
ZeroOrMore(undefined_control)("controls") + block_end
dialog = name_id("block_id") + (Keyword("DIALOGEX") | Keyword("DIALOG")
)("block_type") + block_options + block
string_table = Keyword("STRINGTABLE")("block_type") + block_options + block
menu_item = Keyword("MENUITEM")("block_type") + (
commaSeparatedList("values_") | Keyword("SEPARATOR"))
popup_block = Forward()
popup_block <<= Group(
Keyword("POPUP")("block_type") + Optional(quotedString("caption")) +
block_start + ZeroOrMore(Group(menu_item | popup_block))("elements") +
block_end)("popups*")
menu = name_id("block_id") + \
Keyword("MENU")("block_type") + block_options + \
block_start + ZeroOrMore(popup_block) + block_end
statem = comments ^ precompiler ^ language_definition ^ dialog ^ string_table ^ menu
return statem
