diff options
Diffstat (limited to 'examples/lucene_grammar.py')
| -rw-r--r-- | examples/lucene_grammar.py | 42 |
1 files changed, 25 insertions, 17 deletions
diff --git a/examples/lucene_grammar.py b/examples/lucene_grammar.py index bf92509..ee4633c 100644 --- a/examples/lucene_grammar.py +++ b/examples/lucene_grammar.py @@ -9,19 +9,22 @@ import pyparsing as pp from pyparsing import pyparsing_common as ppc + pp.ParserElement.enablePackrat() -COLON,LBRACK,RBRACK,LBRACE,RBRACE,TILDE,CARAT = map(pp.Literal,":[]{}~^") -LPAR,RPAR = map(pp.Suppress,"()") +COLON, LBRACK, RBRACK, LBRACE, RBRACE, TILDE, CARAT = map(pp.Literal, ":[]{}~^") +LPAR, RPAR = map(pp.Suppress, "()") and_, or_, not_, to_ = map(pp.CaselessKeyword, "AND OR NOT TO".split()) keyword = and_ | or_ | not_ | to_ expression = pp.Forward() -valid_word = pp.Regex(r'([a-zA-Z0-9*_+.-]|\\\\|\\([+\-!(){}\[\]^"~*?:]|\|\||&&))+').setName("word") +valid_word = pp.Regex( + r'([a-zA-Z0-9*_+.-]|\\\\|\\([+\-!(){}\[\]^"~*?:]|\|\||&&))+' +).setName("word") valid_word.setParseAction( - lambda t : t[0].replace('\\\\',chr(127)).replace('\\','').replace(chr(127),'\\') - ) + lambda t: t[0].replace("\\\\", chr(127)).replace("\\", "").replace(chr(127), "\\") +) string = pp.QuotedString('"') @@ -37,24 +40,28 @@ field_name = valid_word().setName("fieldname") incl_range_search = pp.Group(LBRACK - term("lower") + to_ + term("upper") + RBRACK) excl_range_search = pp.Group(LBRACE - term("lower") + to_ + term("upper") + RBRACE) range_search = incl_range_search("incl_range") | excl_range_search("excl_range") -boost = (CARAT - number("boost")) +boost = CARAT - number("boost") string_expr = pp.Group(string + proximity_modifier) | string word_expr = pp.Group(valid_word + fuzzy_modifier) | valid_word -term << (pp.Optional(field_name("field") + COLON) - + (word_expr | string_expr | range_search | pp.Group(LPAR + expression + RPAR)) - + pp.Optional(boost)) -term.setParseAction(lambda t:[t] if 'field' in t or 'boost' in t else None) +term << ( + pp.Optional(field_name("field") + COLON) + + (word_expr | string_expr | range_search | pp.Group(LPAR + expression + RPAR)) + + pp.Optional(boost) +) +term.setParseAction(lambda t: [t] if "field" in t or "boost" in t else None) -expression << pp.infixNotation(term, +expression << pp.infixNotation( + term, [ - (required_modifier | prohibit_modifier, 1, pp.opAssoc.RIGHT), - ((not_ | '!').setParseAction(lambda: "NOT"), 1, pp.opAssoc.RIGHT), - ((and_ | '&&').setParseAction(lambda: "AND"), 2, pp.opAssoc.LEFT), - (pp.Optional(or_ | '||').setParseAction(lambda: "OR"), 2, pp.opAssoc.LEFT), - ]) + (required_modifier | prohibit_modifier, 1, pp.opAssoc.RIGHT), + ((not_ | "!").setParseAction(lambda: "NOT"), 1, pp.opAssoc.RIGHT), + ((and_ | "&&").setParseAction(lambda: "AND"), 2, pp.opAssoc.LEFT), + (pp.Optional(or_ | "||").setParseAction(lambda: "OR"), 2, pp.opAssoc.LEFT), + ], +) -if __name__ == '__main__': +if __name__ == "__main__": # test strings taken from grammar description doc, and TestQueryParser.java tests = r""" @@ -329,4 +336,5 @@ if __name__ == '__main__': if not (success1 and success2): import sys + sys.exit(1) |