#468 -- iter_group_values_lists is considered as None by default (e…

…mpty Synonyms collection)

arekit/common/synonyms/base.py

@@ -5,8 +5,14 @@
 
 class SynonymsCollection(object):
 
-    def __init__(self, iter_group_values_lists, is_read_only, debug):
-        assert(isinstance(iter_group_values_lists, collections.Iterable))
+    def __init__(self, iter_group_values_lists=None, is_read_only=True, debug=False):
+        """ iter_group_values_lists: iterable or None
+            is_read_only: bool
+                whether the relation collection could be expanded or not
+            debug: bool
+                utilized for logging the salient information during usage.
+        """
+        assert(isinstance(iter_group_values_lists, collections.Iterable) or iter_group_values_lists is None)
         assert(isinstance(is_read_only, bool))
         assert(isinstance(debug, bool))
 
@@ -16,7 +22,7 @@ def __init__(self, iter_group_values_lists, is_read_only, debug):
 
         self.__is_read_only = is_read_only
         self.__debug = debug
-        self.__fill(iter_grop_values_lists=iter_group_values_lists)
+        self.__fill(iter_group_values_lists=[] if iter_group_values_lists is None else iter_group_values_lists)
 
     # region properties
 
@@ -102,8 +108,8 @@ def iter_group(self, group_index):
 
     # region private methods
 
-    def __fill(self, iter_grop_values_lists):
-        for group in iter_grop_values_lists:
+    def __fill(self, iter_group_values_lists):
+        for group in iter_group_values_lists:
             self.__process_group(group)
 
     def __process_group(self, group_values_list):

arekit/contrib/source/sentinerel/entities.py

@@ -39,8 +39,7 @@ def read_collection(cls, filename, version, entities_to_ignore=None):
 
         # Since this dataset does not provide the synonyms collection by default,
         # it is necessary to declare an empty collection to populate so in further.
-        synonyms = StemmerBasedSynonymCollection(iter_group_values_lists=[],
-                                                 stemmer=MystemWrapper(),
+        synonyms = StemmerBasedSynonymCollection(stemmer=MystemWrapper(),
                                                  is_read_only=False,
                                                  debug=False)
 

arekit/contrib/utils/pipelines/sources/sentinerel/extract_text_opinions.py

@@ -126,8 +126,7 @@ def create_nolabel_text_opinion_annotator(terms_per_context, dist_in_sents=0, sy
     assert(isinstance(dist_in_sents, int))
 
     if synonyms is None:
-        synonyms = StemmerBasedSynonymCollection(iter_group_values_lists=[],
-                                                 stemmer=MystemWrapper(),
+        synonyms = StemmerBasedSynonymCollection(stemmer=MystemWrapper(),
                                                  is_read_only=False,
                                                  debug=False)
 

arekit/contrib/utils/synonyms/stemmer_based.py

@@ -3,8 +3,15 @@
 
 
 class StemmerBasedSynonymCollection(SynonymsCollection):
-
-    def __init__(self, iter_group_values_lists, stemmer, is_read_only, debug):
+    """ This class represent a synonym collection,
+        in which grouping function is based on the stemmed entity values.
+    """
+
+    def __init__(self, stemmer, iter_group_values_lists=None, is_read_only=True, debug=False):
+        """ iter_group_values_lists: iterable or None
+                describes the original set of values for initialization;
+                If None, then collection is considered as empty.
+        """
         assert(isinstance(stemmer, Stemmer))
         self.__stemmer = stemmer
         super(StemmerBasedSynonymCollection, self).__init__(iter_group_values_lists=iter_group_values_lists,

tests/contrib/source/test_brat_compound_entities.py

@@ -25,8 +25,7 @@ def test(self):
             {"text": self.text, "ind_begin": 0, "ind_end": len(self.text)}
         ]
 
-        synonyms = StemmerBasedSynonymCollection(
-            iter_group_values_lists=[], stemmer=MystemWrapper(), is_read_only=False, debug=False)
+        synonyms = StemmerBasedSynonymCollection(stemmer=MystemWrapper(), is_read_only=False, debug=False)
 
         collection = EntityCollection(
             self.entities,

tests/tutorials/test_tutorial_collection_binding.py

@@ -53,8 +53,7 @@ def __init__(self, contents, value_to_group_id_func):
 
     @classmethod
     def read_collection(cls, filename, version):
-        synonyms = StemmerBasedSynonymCollection(
-            iter_group_values_lists=[], stemmer=MystemWrapper(), is_read_only=False, debug=False)
+        synonyms = StemmerBasedSynonymCollection(stemmer=MystemWrapper(), is_read_only=False, debug=False)
         return FooIOUtils.read_from_zip(
             inner_path=FooIOUtils.get_annotation_innerpath(filename),
             process_func=lambda input_file: cls(

tests/tutorials/test_tutorial_pipeline_text_opinion_annotation.py

@@ -50,8 +50,7 @@ def test(self):
             label_formatter=CustomLabelsFormatter(pos_label_type=PositiveLabel,
                                                   neg_label_type=NegativeLabel))
 
-        synonyms = StemmerBasedSynonymCollection(
-            iter_group_values_lists=[], stemmer=MystemWrapper(), is_read_only=False, debug=False)
+        synonyms = StemmerBasedSynonymCollection(stemmer=MystemWrapper(), is_read_only=False, debug=False)
 
         nolabel_annotator = AlgorithmBasedTextOpinionAnnotator(
             annot_algo=PairBasedOpinionAnnotationAlgorithm(