Deck
Deck

Reputation: 1979

Text generation algorithm

I am looking for an algorithm or some advices to implement it.
Input:

"{1|2} word {3|4}"

Output:

["1 word 3", "1 word 4", "2 word 3", "2 word 4"]

Also, i want to support nested constructions - "{1|2{0|1}}" -> ["1", "20", "21"]
I realize that the question is too general, but don't want to implement the wheel. Maybe you saw similar things.

UPD

from pyparsing import *
from collections import deque

s = u"{1|2} {3|4}"

deque = deque()

def mesh_lists(listOne, listTwo):
    result = []
    for l1 in listOne:
        for l2 in listTwo:
            firstWord = str(l1).strip()
            secondWord = str(l2).strip()
            result.append(" " + firstWord + " " + l2 + " ")
    return result

def action(string, pos, token):
    global deque
    deque.append(list(token[0]))

def processDeque():
    global deque
    while len(deque) > 1:
        l1 = deque.popleft()
        l2 = deque.popleft()
        res = mesh_lists(l1,l2)
        deque.appendleft(res)
    return [x.strip() for x in deque[0]]

_lcurl = Suppress('{')
_rcurl = Suppress('}')
_pipe = Suppress('|')
word = Regex("[^{|}]+")
varBlock = Forward()
entry = word | varBlock
varList = Group(entry + ZeroOrMore(_pipe + entry))
varBlock << (_lcurl + Optional(varList) + _rcurl).setParseAction(action)
template = ZeroOrMore(entry)

res = template.parseString(s)
print processDeque()

It supports only "{||}{||}" constructions. No barewords, no nested constructions.

Upvotes: 2

Views: 451

Answers (2)

PaulMcG
PaulMcG

Reputation: 63739

The pyparsing published examples include a regex inverter.

Upvotes: 0

Chris Martin
Chris Martin

Reputation: 30736

This project looks suitable for you: https://github.com/asciimoo/exrex

Exrex is a tool that generates all matching strings to a given regular expression.

Upvotes: 4

Related Questions