Fix regression test for #615 and remove unnecessary imports

spacy/tests/regression/test_issue615.py

@@ -1,15 +1,17 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import spacy
+from ...matcher import Matcher
-from spacy.attrs import ORTH
+from ...attrs import ORTH
+from ..util import get_doc
 
 
-def merge_phrases(matcher, doc, i, matches):
+def test_issue615(en_tokenizer):
-    '''
+    def merge_phrases(matcher, doc, i, matches):
-    Merge a phrase. We have to be careful here because we'll change the token indices.
+        """Merge a phrase. We have to be careful here because we'll change the
-    To avoid problems, merge all the phrases once we're called on the last match.
+        token indices. To avoid problems, merge all the phrases once we're called
-    '''
+        on the last match."""
+
         if i != len(matches)-1:
             return None
         # Get Span objects
@@ -17,19 +19,16 @@ def merge_phrases(matcher, doc, i, matches):
         for ent_id, label, span in spans:
             span.merge('NNP' if label else span.root.tag_, span.text, doc.vocab.strings[label])
 
-def test_entity_ID_assignment():
+    text = "The golf club is broken"
-    nlp = spacy.en.English()
+    pattern = [{ ORTH: "golf"}, { ORTH: "club"}]
-    text = """The golf club is broken"""
+    label = "Sport_Equipment"
-    doc = nlp(text)
 
-    golf_pattern =     [
+    tokens = en_tokenizer(text)
-            { ORTH: "golf"},
+    doc = get_doc(tokens.vocab, [t.text for t in tokens])
-            { ORTH: "club"}
-        ]
 
-    matcher = spacy.matcher.Matcher(nlp.vocab)
+    matcher = Matcher(doc.vocab)
-    matcher.add_entity('Sport_Equipment', on_match = merge_phrases)
+    matcher.add_entity(label, on_match=merge_phrases)
-    matcher.add_pattern("Sport_Equipment", golf_pattern, label = 'Sport_Equipment')
+    matcher.add_pattern(label, pattern, label=label)
 
     match = matcher(doc)
     entities = list(doc.ents)