def test7(self):
# def diff(quotedText, unquotedText, baseName, goldAlreadyTokenized):
res = ftd.diff(self.text7t, self.text7g, self.text7, True)
self.assertEqual(res["MERGE"], [(self.text7, 19, 20)])
def test5(self):
# def diff(quotedText, unquotedText, baseName, goldAlreadyTokenized):
res = ftd.diff(self.text5t, self.text5g, self.text5, True)
self.assertEqual(res["MERGE"], [])
def test6(self):
# def diff(quotedText, unquotedText, baseName, goldAlreadyTokenized):
res = ftd.diff(self.text6t, self.text6g, self.text6, True)
#self.assertEqual(res["MERGE"], [(self.text6, 29, 30)])
self.assertEqual(res["MERGE"], [(self.text6, 28, 29)])
def test4a(self):
res = ftd.diff(self.text4qs, self.text4u, self.text4, True)
# Does not matter, crashes...
self.assertEqual(res["MERGE"], [])
def test4(self):
# fails because here the gold is split unexpectedly
res = ftd.diff(self.text4q, self.text4u, self.text4, False)
# Does not matter, crashes...
self.assertEqual(res["MERGE"], [])
def test3(self):
# tests whether actual differences between sentence splits are recognized
# so we don't just check on number of sentence
res = ftd.diff(self.text3_1, self.text3_2, self.text3, False)
self.assertNotEqual(res["MERGE"], [])
def test2(self):
res = ftd.diff(self.text2q, self.text2u, self.text2, False)
self.assertEqual(res["MERGE"], [(self.text2, 39, 40)])
def test1(self):
res = ftd.diff(self.text1q, self.text1u, self.text1, False)
self.assertEqual(res["MERGE"], [(self.text1, 27, 28)])
