[Issue-2670] Bug fix: Initialize doc_no2 because it is not set when c…

…orpus' is empty (#2672)

* [Issue-2670] Bug fix: Initialize doc_no2 because it is not set when 'corpus' is empty

* [Issue-2670] Add: unittests should fail on invalid input (generator and empty corpus)

* [Issue-2670] Add: Fix unittest for generator

* [Issue-2670] Fix unittest tox:flake8 errors

* [Issue-2670] Fix: empty corpus def in unittest

* [Issue-2670] Fix: empty corpus and generator unittests

* [Issue-2670] Fix: empty corpus and generator unittests

gensim/models/logentropy_model.py

@@ -108,6 +108,9 @@ def initialize(self, corpus):
             self.n_docs, len(glob_freq), self.n_words
         )
         logger.debug('iterating over corpus')
+
+        # initialize doc_no2 index in case corpus is empty
+        doc_no2 = 0
         for doc_no2, bow in enumerate(corpus):
             for key, freq in bow:
                 p = (float(freq) / glob_freq[key]) * math.log(float(freq) / glob_freq[key])

gensim/test/test_logentropy_model.py

@@ -19,9 +19,23 @@
 
 
 class TestLogEntropyModel(unittest.TestCase):
+    TEST_CORPUS = [[(1, 1.0)], [], [(0, 0.5), (2, 1.0)], []]
+
     def setUp(self):
         self.corpus_small = MmCorpus(datapath('test_corpus_small.mm'))
         self.corpus_ok = MmCorpus(datapath('test_corpus_ok.mm'))
+        self.corpus_empty = []
+
+    def test_generator_fail(self):
+        """Test creating a model using a generator as input; should fail."""
+        def get_generator(test_corpus=TestLogEntropyModel.TEST_CORPUS):
+            for test_doc in test_corpus:
+                yield test_doc
+        self.assertRaises(ValueError, logentropy_model.LogEntropyModel, corpus=get_generator())
+
+    def test_empty_fail(self):
+        """Test creating a model using an empty input; should fail."""
+        self.assertRaises(ValueError, logentropy_model.LogEntropyModel, corpus=self.corpus_empty)
 
     def testTransform(self):
         # create the transformation model