使用 pyparsing 使用 `delimitedList` 设置最大出现次数

Question

pyparsing provides a helper function, delimitedList , that matches a sequence of one or more expressions , separated with a delimiter :

delimitedList(expr, delim=',', combine=False)

How can this be used to match a sequence of expressions, where each expression may occur zero or one times?

For example, to match "foo", "bar, "baz" I took a bottom-up approach a created a token for each word:

import pyparsing as pp

dbl_quote = pp.Suppress('"')

foo = dbl_quote + pp.Literal('foo') + dbl_quote
bar = dbl_quote + pp.Literal('bar') + dbl_quote
baz = dbl_quote + pp.Literal('baz') + dbl_quote

I want to create an expression that matches:

zero or one occurrences of "foo" , zero or one occurrences of "bar" , zero or one occurrences of "baz"

... in any order . Examples of valid input:

• "foo", "bar", "baz"
• "baz", "bar", "foo", // Order is unimportant
• "bar", "baz" // Zero occurrences allowed
• "baz"
• // Zero occurrences of all tokens

Examples of invalid input:

• "notfoo", "notbar", "notbaz"
• "foo", "foo", "bar", "baz" // Two occurrences of foo
• "foo" "bar", "baz" // Missing comma
• "foo" "bar", "baz", // Trailing comma

I gravitated towards delimitedList because my input is a comma delimited list, but now I feel this function is working against me rather than for me.

import pyparsing as pp

dbl_quote = pp.Suppress('"')

foo = dbl_quote + pp.Literal('foo') + dbl_quote
bar = dbl_quote + pp.Literal('bar') + dbl_quote
baz = dbl_quote + pp.Literal('baz') + dbl_quote



# This is NOT what I want because it allows tokens
# to occur more than once.
foobarbaz = pp.delimitedList(foo | bar | baz)



if __name__ == "__main__":
    TEST = '"foo", "bar", "baz"'
    results = foobarbaz.parseString(TEST)
    results.pprint()

Answer 1

Ordinarily, when I see "in any order" as part of a grammar, my first thought is to use Each , which you can create with the & operator:

undelimited_foo_bar_baz = foo & bar & baz

This parser would parse foo , bar , and baz in any order. If you wanted them to be optional, then simply wrap them in Optional:

undelimited_foo_bar_baz = Optional(foo) & Optional(bar) & Optional(baz)

But the intervening commas in your input make this kind of messy, so as a fallback, you can stick with the delimitedList (which will strip out the commas) add a condition parse action to get run after the list is parsed, to verify that only one of each of the matched items was present:

from collections import Counter
def no_more_than_one_of_any(t):
    return all(freq == 1 for freq in Counter(t.asList()).values())
foobarbaz.addCondition(no_more_than_one_of_any, message="duplicate item found in list")

if __name__ == "__main__":
    tests = '''\
    "foo"
    "bar"
    "baz"
    "foo", "baz"
    "foo", "bar", "baz"
    "foo", "bar", "baz", "foo"
    '''
    foobarbaz.runTests(tests)

Prints:

"foo"
['foo']

"bar"
['bar']

"baz"
['baz']

"foo", "baz"
['foo', 'baz']

"foo", "bar", "baz"
['foo', 'bar', 'baz']

"foo", "bar", "baz", "foo"
^
FAIL: duplicate item found in list, found '"'  (at char 0), (line:1, col:1)