pyparsing matching any combination of specified Literals

Example: I have the literals "alpha", "beta", "gamma". How do I make pyparsing parse the following inputs:

alpha
alpha|beta
beta|alpha|gamma

The given input can be constructed by using one or more non-repeating literals from a given set, separated by "|". Advice o开发者_运维技巧n setting up pyparsing will be appreciated.

Use the '&' operator for Each, instead of '+ or '|'. If you must have all, but in unpredicatable order use:

Literal('alpha') & 'beta' & 'gamma'

If some may be missing, but each used at most once, then use Optionals:

Optional('alpha') & Optional('beta') & Optional('gamma')

Oops, I forgot the '|' delimiters. One lenient parser would be to use a delimitedList:

delimitedList(oneOf("alpha beta gamma"), '|')

This would allow any or all of your choices, but does not guard against duplicates. May be simplest to use a parse action:

itemlist = delimitedList(oneOf("alpha beta gamma"), '|')
def ensureNoDuplicates(tokens):
    if len(set(tokens)) != len(tokens):
        raise ParseException("duplicate list entries found")
itemlist.setParseAction(ensureNoDuplicates)

This feels like the simplest approach to me.

EDIT:

Recent versions of pyparsing have introduced parse-time conditions to make this kind of parse action easier to write:

itemlist = delimitedList(oneOf("alpha beta gamma"), '|')
itemlist.addCondition(lambda tokens: len(set(tokens)) == len(tokens),
                      "duplicate list entries found")