import os
import re
from twisted.plugin import IPlugin
from tmat.tokenizer import TokenizerPlugin
from tmat.myparser import ParserPlugin
from zope.interface import Interface, Attribute, implements
from sets import Set
from xml.dom.minidom import getDOMImplementation
import random

class WhitespaceTokenizer(object):
    implements(IPlugin, TokenizerPlugin)
    running = False
    displayName = "Whitespace Tokenizer"
    preferences = {}
    values = {}
    def __init__(self):
	self.running=True
    def splitText(self, text):
	return text.split(" ")
    def tokenize(self, text):
	splitWords=self.splitText(text) 
	i=0
	retList=[]
	for token in splitWords:
	    i=text.find(token,i)
	    retList.append((i,i+len(token),True))
	    i += len(token)
	return retList
    def start(self):
	self.running=True
    def stop(self):
	self.running=False
    def version(self):
	return "I don't have a version yet"

class DefaultParser(object):
    implements(IPlugin, ParserPlugin)
    running = False
    displayName = "Default Parser"
    preferences = {}
    values = {}
    def initialize(self):
	pass
    def start(self):
	self.running=True
    def stop(self):
	self.running=False
    def version(self):
	return ""
    def uninitialize(self):
	pass
    def parse(self, wordforms, progressBar=None):
	random.seed()
	retDict={}
	for word in wordforms:
	    retDict[word]=[]
	    for i in range(random.randint(1,5)):
		retDict[word].append({"morpheme":"%s-%s" % (word,i),"gloss":"%s/%s" % (word,i)})
	return retDict

x = WhitespaceTokenizer()
y = DefaultParser()