def test_writing(self):
# First read in stuff.
mnist_builder = mnist.MNIST(
data_dir=tempfile.mkdtemp(dir=self.get_temp_dir()))
info = dataset_info.DatasetInfo(
builder=mnist_builder, features=mnist_builder.info.features)
info.read_from_directory(_INFO_DIR)
# Read the json file into a string.
with tf.io.gfile.GFile(info._dataset_info_path(_INFO_DIR)) as f:
existing_json = json.load(f)
# Now write to a temp directory.
with testing.tmp_dir(self.get_temp_dir()) as tmp_dir:
info.write_to_directory(tmp_dir)
# Read the newly written json file into a string.
with tf.io.gfile.GFile(info._dataset_info_path(tmp_dir)) as f:
new_json = json.load(f)
# Read the newly written LICENSE file into a string.
with tf.io.gfile.GFile(info._license_path(tmp_dir)) as f:
license_ = f.read()
# Assert what was read and then written and read again is the same.
self.assertEqual(existing_json, new_json)
# Assert correct license was written.
self.assertEqual(existing_json["redistributionInfo"]["license"], license_)
if six.PY3:
# Only test on Python 3 to avoid u'' formatting issues
self.assertEqual(repr(info), INFO_STR)
def test_writing(self):
# First read in stuff.
mnist_builder = mnist.MNIST(data_dir=tempfile.mkdtemp(
dir=self.get_temp_dir()))
info = dataset_info.DatasetInfo(builder=mnist_builder,
features=mnist_builder.info.features)
info.read_from_directory(_INFO_DIR)
# Read the json file into a string.
with tf.io.gfile.GFile(info._dataset_info_path(_INFO_DIR)) as f:
existing_json = json.load(f)
# Now write to a temp directory.
with testing.tmp_dir(self.get_temp_dir()) as tmp_dir:
info.write_to_directory(tmp_dir)
# Read the newly written json file into a string.
with tf.io.gfile.GFile(info._dataset_info_path(tmp_dir)) as f:
new_json = json.load(f)
# Read the newly written LICENSE file into a string.
with tf.io.gfile.GFile(info._license_path(tmp_dir)) as f:
license_ = f.read()
# Assert what was read and then written and read again is the same.
self.assertEqual(existing_json, new_json)
# Assert correct license was written.
self.assertEqual(existing_json["redistributionInfo"]["license"],
license_)
# Do not check the full string as it display the generated path.
self.assertEqual(_INFO_STR % mnist_builder.data_dir, repr(info))
self.assertIn("'test': <SplitInfo num_examples=", repr(info))
def test_reading_from_gcs_bucket(self):
# The base TestCase prevents GCS access, so we explicitly ask it to restore
# access here.
with self.gcs_access():
mnist_builder = mnist.MNIST(data_dir=tempfile.mkdtemp(
dir=self.get_temp_dir()))
info = dataset_info.DatasetInfo(builder=mnist_builder)
info = mnist_builder.info
# A nominal check to see if we read it.
self.assertTrue(info.initialized)
self.assertEqual(10000, info.splits["test"].num_examples)
def test_str_smoke(self):
info = mnist.MNIST(data_dir="/tmp/some_dummy_dir").info
_ = str(info)
def setUpClass(cls):
super(DatasetInfoTest, cls).setUpClass()
cls._tfds_tmp_dir = testing.make_tmp_dir()
cls._builder = mnist.MNIST(data_dir=cls._tfds_tmp_dir)
