From cccc1bfd57e1ec05fbb997346cf54e906837b399 Mon Sep 17 00:00:00 2001
From: Yige Xu <xuyige1996@gmail.com>
Date: Tue, 17 Sep 2019 19:37:37 +0800
Subject: [PATCH] add fitlog and spacy while testing

---
 .travis.yml                   |  2 ++
 test/io/pipe/test_matching.py | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 36 insertions(+)

diff --git a/.travis.yml b/.travis.yml
index bd7a34f5..0770d4e7 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -4,10 +4,12 @@ python:
 # command to install dependencies
 install:
   - pip install --quiet -r requirements.txt
+  - pip install --quiet fitlog
   - pip install pytest>=3.6
   - pip install pytest-cov
 # command to run tests
 script:
+  - python -m spacy download en
   - pytest --cov=fastNLP test/
 
 after_success:
diff --git a/test/io/pipe/test_matching.py b/test/io/pipe/test_matching.py
index 785d44bb..a901bc78 100644
--- a/test/io/pipe/test_matching.py
+++ b/test/io/pipe/test_matching.py
@@ -72,3 +72,37 @@ class TestRunMatchingPipe(unittest.TestCase):
             for x, y in zip(vocab, data_bundle1.iter_vocabs()):
                 name, vocabs = y
                 self.assertEqual(x + 1 if name == 'words' else x, len(vocabs))
+
+    def test_spacy(self):
+        data_set_dict = {
+            'RTE': ('test/data_for_tests/io/RTE', RTEPipe, RTEBertPipe, (5, 5, 5), (425, 2)),
+            }
+        for k, v in data_set_dict.items():
+            path, pipe1, pipe2, data_set, vocab = v
+
+            with self.assertWarns(Warning):
+                data_bundle1 = pipe1(tokenizer='spacy').process_from_file(path)
+                data_bundle2 = pipe2(tokenizer='spacy').process_from_file(path)
+
+            self.assertTrue(isinstance(data_bundle1, DataBundle))
+            self.assertEqual(len(data_set), data_bundle1.num_dataset)
+            print(k)
+            print(data_bundle1)
+            print(data_bundle2)
+            for x, y in zip(data_set, data_bundle1.iter_datasets()):
+                name, dataset = y
+                self.assertEqual(x, len(dataset))
+            self.assertEqual(len(data_set), data_bundle2.num_dataset)
+            for x, y in zip(data_set, data_bundle2.iter_datasets()):
+                name, dataset = y
+                self.assertEqual(x, len(dataset))
+
+            self.assertEqual(len(vocab), data_bundle1.num_vocab)
+            for x, y in zip(vocab, data_bundle1.iter_vocabs()):
+                name, vocabs = y
+                self.assertEqual(x, len(vocabs))
+            self.assertEqual(len(vocab), data_bundle2.num_vocab)
+            for x, y in zip(vocab, data_bundle1.iter_vocabs()):
+                name, vocabs = y
+                self.assertEqual(x + 1 if name == 'words' else x, len(vocabs))
+