[Zope-Checkins] CVS: Zope/lib/python/Products/PluginIndexes/TextIndexNG/tests - testSimilarityLexicon.py:1.1.2.1 testProximityLexicon.py:NONE

[Andreas Jung]

Update of /cvs-repository/Zope/lib/python/Products/PluginIndexes/TextIndexNG/tests
In directory cvs.zope.org:/tmp/cvs-serv28482

Added Files:
      Tag: ajung-textindexng-branch
	testSimilarityLexicon.py 
Removed Files:
      Tag: ajung-textindexng-branch
	testProximityLexicon.py 
Log Message:
renamed all 'Proximity' stuff to 'Similarity'


=== Added File Zope/lib/python/Products/PluginIndexes/TextIndexNG/tests/testSimilarityLexicon.py ===
##############################################################################
#
# Copyright (c) 2001 Zope Corporation and Contributors. All Rights Reserved.
# 
# This software is subject to the provisions of the Zope Public License,
# Version 2.0 (ZPL).  A copy of the ZPL should accompany this distribution.
# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
# FOR A PARTICULAR PURPOSE
# 
##############################################################################

import sys, os, unittest

from Products.PluginIndexes.TextIndexNG.ProximityLexicon import ProximityLexicon
from BTrees.IIBTree import IISet, difference
import Proximity

class Tests(unittest.TestCase):

    _metaphoneAscii= {
        'brown' : ['braun','Brun','braune','browne'],
        'quick' : ['qicke','quicke','quicky','kik','kuik'],
    }

    _metaphoneLatin1= {
        'alle' : ['ale','ahle','aale','al'],
        'einen' : ['eihnen','einnen','eihnnen'],
        'zerstört' : ['zerstoert','zerstörd','zersdörd','zerzdoerd'],
    }

    _soundexAscii = {
        'brown' : ['braun','Brun','braune','browne'],
        'quick' : ['qicke','quicke','quicky','qwik'],
    }

    _soundexLatin1 = {
        'alle' : ['ale','ahle','aale','al'],
        'einen' : ['eihnen','einnen','eihnnen'],
        'zerstört' : ['zerstoert','zerstörd','zersdörd','zerzdoerd'],
    }

    def doTest(self,s, algorithm):
        """ """
    
        self._lexicon  = ProximityLexicon(algorithm)    

        words = s.split()

        self.assertEqual(len(self._lexicon),0)

        self.WIDS = {}
        for word in words:
            self.WIDS[word] = self._lexicon.getWordId(word)

        self.assertEqual(len(self.WIDS) , len(self._lexicon._lexicon))            


        for word,wid in self.WIDS.items():

            set = self._lexicon.get(word)

            diff = difference(IISet([wid]), set)
            self.assertEqual( len(diff),0, (word,wid))


    def algorithmTest(self,d):

        for key,words in d.items():

            for word in words:

                i1 = self._lexicon.get(key)
                i2 = self._lexicon.get(word)

                self.assertEqual( len(difference(i1,i2)), 0, ( key,word))


    def testSoundexAscii(self):
        """ Soundex Ascii test """

        s = 'the quick brown fox jumps over the lazy dog'
        self.doTest(s,'soundex')
        self.algorithmTest(self._soundexAscii)


    def testSoundexLatin1(self):
        """ Soundex Latin1 test """

        s = ( 'auf einen alle Flugplätze der Taliban zerstört worden')
        self.doTest(s,'soundex')
        self.algorithmTest(self._soundexLatin1)


    def testSoundexUnicode(self):
        """ Soundex unicode test """

        s = unicode('auf einen alle Flugplätze der Taliban zerstört worden',
                'latin1')
        self.doTest(s,'soundex')
        self.algorithmTest(self._soundexLatin1)


    def testMetaphoneAscii(self):
        """ Soundex Ascii test """

        s = 'the quick brown fox jumps over the lazy dog' 
        self.doTest(s,'metaphone')
        self.algorithmTest(self._metaphoneAscii)


    def testMetaphoneLatin1(self):
        """ Soundex Latin1 test """

        s = 'auf einen alle Flugplätze der Taliban zerstört worden'
        self.doTest(s,'metaphone')
        self.algorithmTest(self._metaphoneLatin1)


    def testMetaphoneUnicode(self):
        """ Metaphone Unicode test """

        s = unicode('auf einen alle Flugplätze der Taliban zerstört worden',
                'latin1')
        self.doTest(s,'metaphone')

    def printLexicon(self):

        for k,v in self._lexicon._lexicon.items():
            print k.encode('latin1'),v
        
       

def test_suite():
   return unittest.makeSuite(Tests)

def main():
   unittest.TextTestRunner().run(test_suite())

def debug():
   test_suite().debug()

def pdebug():
    import pdb
    pdb.run('debug()')
   
if __name__=='__main__':
   if len(sys.argv) > 1:
      globals()[sys.argv[1]]()
   else:
      main()


=== Removed File Zope/lib/python/Products/PluginIndexes/TextIndexNG/tests/testProximityLexicon.py ===