Extracting information from ParseResults in pyparsing - parsing

Question

I'm discovering the pyparsing module which is really cool. I'm trying to parse a set of simple boolean expressions in which some identifiers (foo/bar/bla and zoo) are compared to numeric values. The parser is used to check that user expression is correct but I would like also to get the name of the identifiers used in the expression (i.e which combination of foo/bar/bla and zoo was used). I can't get a simple way to do it.
In the example below foo and bar are used in the expression. But how can I get this information ?
Best
from pyparsing import oneOf
from pyparsing import Group
from pyparsing import Regex
from pyparsing import operatorPrecedence
from pyparsing import opAssoc
from pyparsing import Literal
from pyparsing import Word
from pyparsing import nums
from pyparsing import Combine
from pyparsing import Optional
from pyparsing import CaselessLiteral
from pyparsing import alphanums
from pyparsing import quotedString
from pyparsing import Forward
lparen = Literal("(")
rparen = Literal(")")
and_operator = CaselessLiteral("and")
or_operator = CaselessLiteral("or")
comparison_operator = oneOf(['==','!=','>','>=','<', '<='])
point = Literal('.')
e = CaselessLiteral('E')
plusorminus = Literal('+') | Literal('-')
number = Word(nums)
integer = Combine( Optional(plusorminus) + number )
float_nb = Combine( integer +
Optional( point + Optional(number) ) +
Optional( e + integer ))
value = float_nb
value.resultsName = 'value'
identifier = oneOf(['foo','bar', 'bla', 'zoo'], caseless=False)
identifier.resultsName = 'key'
group_1 = Group(identifier + comparison_operator + value)
group_2 = Group(value + comparison_operator + identifier)
comparison = group_1 | group_2
boolean_expr = operatorPrecedence(
comparison,
[(and_operator, 2, opAssoc.LEFT),
(or_operator, 2, opAssoc.LEFT)])
boolean_expr_par = "(" + boolean_expr + ")"
expression = Forward()
expression << boolean_expr | boolean_expr_par
exp = expression.parseString('2.5 > foo and (3 < bar or (foo > 10 and bar < 3)) ' , parseAll=True)
# Now how can I get the 'identifiers used in exp' ?

Answer 1

I had a similar problem.
I used the 'setParseAction' method in your 'identifier' parser to set a function recording each occurrence of the token matching 'identifier'. Then I print the recorded items:
I declare :
idSet = set()
def recordID(tokens):
idSet.add(tokens[0])
return
I modify your 'identifier' parser as follows:
identifier = oneOf(['foo','bar', 'bla', 'zoo'], caseless=False).setParseAction(recordID)
At the end of you script I print 'idSet':
exp = expression.parseString('2.5 > foo and (3 < bar or (foo > 10 and bar < 3)) ' , parseAll=True)
print(idSet)
It gives the following outcome:
{'foo', 'bar'}