def _make_simple_parser():
ParserElement.setDefaultWhitespaceChars(" \n\t\r'")
wordchars = printables
for specialchar in ':+-"':
wordchars = wordchars.replace(specialchar, "")
wordtoken = Combine(Word(wordchars) + ZeroOrMore("." + Word(wordchars)))
# A word-like thing
generalWord = Group(wordtoken).setResultsName("Word")
# A quoted phrase
quotedPhrase = Group(Suppress('"') + CharsNotIn('"') + Suppress('"')).setResultsName("Quotes")
# Units of content
fieldableUnit = quotedPhrase | generalWord
fieldedUnit = Group(Word(alphanums) + Suppress(':') + fieldableUnit).setResultsName("Field")
unit = fieldedUnit | fieldableUnit
# A unit may be "not"-ed.
operatorNot = Group(Suppress(Literal("-")) + unit).setResultsName("Not")
# A unit may be required
operatorReqd = Group(Suppress(Literal("+")) + unit).setResultsName("Required")
generalUnit = operatorNot | operatorReqd | unit
expression = (OneOrMore(generalUnit) | Empty())
toplevel = Group(expression).setResultsName("Toplevel") + StringEnd()
return toplevel.parseString
def _make_simple_parser():
ParserElement.setDefaultWhitespaceChars(" \n\t\r'")
wordToken = Regex(r"(\w|/)+(\.?(\w|\-|/)+)*", re.UNICODE)
# A word-like thing
generalWord = Group(wordToken).setResultsName("Word")
# A quoted phrase
quotedPhrase = Group(Suppress('"') + CharsNotIn('"') +
Suppress('"')).setResultsName("Quotes")
# Units of content
fieldableUnit = quotedPhrase | generalWord
fieldedUnit = Group(Word(alphanums) + Suppress(':') +
fieldableUnit).setResultsName("Field")
unit = fieldedUnit | fieldableUnit
# A unit may be "not"-ed.
operatorNot = Group(Suppress(Literal("-")) + unit).setResultsName("Not")
# A unit may be required
operatorReqd = Group(Suppress(Literal("+")) +
unit).setResultsName("Required")
generalUnit = operatorNot | operatorReqd | unit
expression = (OneOrMore(generalUnit) | Empty())
toplevel = Group(expression).setResultsName("Toplevel") + StringEnd()
return toplevel.parseString
def _make_default_parser():
ParserElement.setDefaultWhitespaceChars(" \n\t\r'")
#wordToken = Word(self.wordChars)
wordToken = Regex(r"(\w|/)+(\.?(\w|\-|/)+)*", re.UNICODE)
# A plain old word.
plainWord = Group(wordToken).setResultsName("Word")
# A word ending in a star (e.g. 'render*'), indicating that
# the search should do prefix expansion.
prefixWord = Group(Combine(wordToken + Suppress('*'))).setResultsName("Prefix")
# A wildcard word containing * or ?.
wildcard = Group(Regex(r"\w*(?:[\?\*]\w*)+")).setResultsName("Wildcard")
# A range of terms
range = Group(plainWord + Suppress("..") + plainWord).setResultsName("Range")
# A word-like thing
generalWord = range | prefixWord | wildcard | plainWord
# A quoted phrase
quotedPhrase = Group(Suppress('"') + CharsNotIn('"') + Suppress('"')).setResultsName("Quotes")
expression = Forward()
# Parentheses can enclose (group) any expression
parenthetical = Group((Suppress("(") + expression + Suppress(")"))).setResultsName("Group")
boostableUnit = quotedPhrase | generalWord
boostedUnit = Group(boostableUnit + Suppress("^") + Word("0123456789", ".0123456789")).setResultsName("Boost")
# The user can flag that a parenthetical group, quoted phrase, or word
# should be searched in a particular field by prepending 'fn:', where fn is
# the name of the field.
fieldableUnit = parenthetical | boostedUnit | boostableUnit
fieldedUnit = Group(Word(alphanums) + Suppress(':') + fieldableUnit).setResultsName("Field")
# Units of content
unit = fieldedUnit | fieldableUnit
# A unit may be "not"-ed.
operatorNot = Group(Suppress(Keyword("not", caseless=True)) + unit).setResultsName("Not")
generalUnit = operatorNot | unit
andToken = Keyword("and", caseless=True)
orToken = Keyword("or", caseless=True)
operatorAnd = Group(generalUnit + Suppress(andToken) + expression).setResultsName("And")
operatorOr = Group(generalUnit + Suppress(orToken) + expression).setResultsName("Or")
expression << (OneOrMore(operatorAnd | operatorOr | generalUnit) | Empty())
toplevel = Group(expression).setResultsName("Toplevel") + StringEnd()
return toplevel.parseString
def _make_default_parser():
escapechar = "\\"
#wordchars = printables
#for specialchar in '*?^():"{}[] ' + escapechar:
# wordchars = wordchars.replace(specialchar, "")
#wordtext = Word(wordchars)
wordtext = CharsNotIn('\\*?^():"{}[] ')
escape = Suppress(escapechar) + (Word(printables, exact=1) | White(exact=1))
wordtoken = Combine(OneOrMore(wordtext | escape))
# A plain old word.
plainWord = Group(wordtoken).setResultsName("Word")
# A wildcard word containing * or ?.
wildchars = Word("?*")
# Start with word chars and then have wild chars mixed in
wildmixed = wordtoken + OneOrMore(wildchars + Optional(wordtoken))
# Or, start with wildchars, and then either a mixture of word and wild chars, or the next token
wildstart = wildchars + (OneOrMore(wordtoken + Optional(wildchars)) | FollowedBy(White() | StringEnd()))
wildcard = Group(Combine(wildmixed | wildstart)).setResultsName("Wildcard")
# A range of terms
startfence = Literal("[") | Literal("{")
endfence = Literal("]") | Literal("}")
rangeitem = QuotedString('"') | wordtoken
openstartrange = Group(Empty()) + Suppress(Keyword("TO") + White()) + Group(rangeitem)
openendrange = Group(rangeitem) + Suppress(White() + Keyword("TO")) + Group(Empty())
normalrange = Group(rangeitem) + Suppress(White() + Keyword("TO") + White()) + Group(rangeitem)
range = Group(startfence + (normalrange | openstartrange | openendrange) + endfence).setResultsName("Range")
# rangeitem = QuotedString('"') | wordtoken
# rangestartitem = Group((rangeitem + Suppress(White())) | Empty()).setResultsName("rangestart")
# rangeenditem = Group((Suppress(White()) + rangeitem) | Empty()).setResultsName("rangeend")
# rangestart = (Literal("{") | Literal("[")) + rangestartitem
# rangeend = rangeenditem + (Literal("}") | Literal("]"))
# range = Group(rangestart + Suppress(Literal("TO")) + rangeend).setResultsName("Range")
# A word-like thing
generalWord = range | wildcard | plainWord
# A quoted phrase
quotedPhrase = Group(QuotedString('"')).setResultsName("Quotes")
expression = Forward()
# Parentheses can enclose (group) any expression
parenthetical = Group((Suppress("(") + expression + Suppress(")"))).setResultsName("Group")
boostableUnit = generalWord | quotedPhrase
boostedUnit = Group(boostableUnit + Suppress("^") + Word("0123456789", ".0123456789")).setResultsName("Boost")
# The user can flag that a parenthetical group, quoted phrase, or word
# should be searched in a particular field by prepending 'fn:', where fn is
# the name of the field.
fieldableUnit = parenthetical | boostedUnit | boostableUnit
fieldedUnit = Group(Word(alphanums + "_") + Suppress(':') + fieldableUnit).setResultsName("Field")
# Units of content
unit = fieldedUnit | fieldableUnit
# A unit may be "not"-ed.
operatorNot = Group(Suppress(Keyword("not", caseless=True)) + Suppress(White()) + unit).setResultsName("Not")
generalUnit = operatorNot | unit
andToken = Keyword("AND", caseless=False)
orToken = Keyword("OR", caseless=False)
andNotToken = Keyword("ANDNOT", caseless=False)
operatorAnd = Group(generalUnit + Suppress(White()) + Suppress(andToken) + Suppress(White()) + expression).setResultsName("And")
operatorOr = Group(generalUnit + Suppress(White()) + Suppress(orToken) + Suppress(White()) + expression).setResultsName("Or")
operatorAndNot = Group(unit + Suppress(White()) + Suppress(andNotToken) + Suppress(White()) + unit).setResultsName("AndNot")
expression << (OneOrMore(operatorAnd | operatorOr | operatorAndNot | generalUnit | Suppress(White())) | Empty())
toplevel = Group(expression).setResultsName("Toplevel") + StringEnd()
return toplevel.parseString
def _make_default_parser():
escapechar = "\\"
wordchars = printables
for specialchar in '*?^():."' + escapechar:
wordchars = wordchars.replace(specialchar, "")
wordtext = Combine(Word(wordchars) + ZeroOrMore("." + Word(wordchars)))
escape = Suppress(escapechar) + (Word(printables, exact=1)
| White(exact=1))
wordtoken = Combine(OneOrMore(wordtext | escape))
# A plain old word.
plainWord = Group(wordtoken).setResultsName("Word")
# A word ending in a star (e.g. 'render*'), indicating that
# the search should do prefix expansion.
prefixWord = Group(Combine(wordtoken +
Suppress('*'))).setResultsName("Prefix")
# A wildcard word containing * or ?.
wildcard = Group(Regex(r"\w*(?:[\?\*]\w*)+")).setResultsName("Wildcard")
# A range of terms
range = Group(plainWord +
Suppress(Optional(White()) + ".." + Optional(White())) +
plainWord).setResultsName("Range")
# A word-like thing
generalWord = range | prefixWord | wildcard | plainWord
# A quoted phrase
quotedPhrase = Group(Suppress('"') + CharsNotIn('"') +
Suppress('"')).setResultsName("Quotes")
expression = Forward()
# Parentheses can enclose (group) any expression
parenthetical = Group(
(Suppress("(") + expression + Suppress(")"))).setResultsName("Group")
boostableUnit = quotedPhrase | generalWord
boostedUnit = Group(boostableUnit + Suppress("^") +
Word("0123456789", ".0123456789")).setResultsName(
"Boost")
# The user can flag that a parenthetical group, quoted phrase, or word
# should be searched in a particular field by prepending 'fn:', where fn is
# the name of the field.
fieldableUnit = parenthetical | boostedUnit | boostableUnit
fieldedUnit = Group(Word(alphanums + "_") + Suppress(':') +
fieldableUnit).setResultsName("Field")
# Units of content
unit = fieldedUnit | fieldableUnit
# A unit may be "not"-ed.
operatorNot = Group(
Suppress(Keyword("not", caseless=True)) + Suppress(White()) +
unit).setResultsName("Not")
generalUnit = operatorNot | unit
andToken = Keyword("and", caseless=True)
orToken = Keyword("or", caseless=True)
operatorAnd = Group(generalUnit + Suppress(White()) + Suppress(andToken) +
Suppress(White()) + expression).setResultsName("And")
operatorOr = Group(generalUnit + Suppress(White()) + Suppress(orToken) +
Suppress(White()) + expression).setResultsName("Or")
expression << (OneOrMore(operatorAnd | operatorOr | generalUnit
| Suppress(White())) | Empty())
toplevel = Group(expression).setResultsName("Toplevel") + StringEnd()
return toplevel.parseString