def _create_non_empty_dir_(self, path):
_dir = HDFS(path)
_dir.create_directory()
self.assertTrue(_dir.exists(), "source directory not found")
for i in range(5):
_file = HDFS(os.path.join(path, str(uuid.uuid4())))
_file.create(directory=(i % 2 == 0))
self.assertTrue(_file.exists(), "File was not created")
return _dir
def _create_non_empty_dir_(self, path):
_dir = HDFS(path)
_dir.create_directory()
self.assertTrue(_dir.exists(), "source directory not found")
for i in range(5):
_file = HDFS(os.path.join(path, str(uuid.uuid4())))
_file.create(directory=(i % 2 == 0))
self.assertTrue(_file.exists(), "File was not created")
return _dir
def test_create_directory(self):
new_dir = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
self.assertFalse(new_dir.exists(), "Directory is already exists")
try:
new_dir.create_directory()
self.assertTrue(new_dir.exists(), "Directory was not created")
self.assertTrue(new_dir.is_directory())
finally:
new_dir.delete(recursive=True)
self.assertFalse(new_dir.exists(), "Directory was not removed")
def test_create_directory(self):
new_dir = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
self.assertFalse(new_dir.exists(), "Directory is already exists")
try:
new_dir.create_directory()
self.assertTrue(new_dir.exists(), "Directory was not created")
self.assertTrue(new_dir.is_directory())
finally:
new_dir.delete(recursive=True)
self.assertFalse(new_dir.exists(), "Directory was not removed")
def should_create_directory_recursively(self):
_base_dir = os.path.join("/tmp", str(uuid.uuid4()))
_path = os.path.join(_base_dir, str(uuid.uuid4()), str(uuid.uuid4()))
_dir = HDFS(_path)
self.assertFalse(_dir.exists(), "Folder is already exists")
try:
_dir.create_directory(recursive=True)
self.assertTrue(_dir.exists(), "Folder was not created")
self.assertTrue(_dir.is_directory(), "New file should be a directory")
finally:
HDFS(_base_dir).delete_directory()
self.assertFalse(_dir.exists(), "File was not removed")
self.assertFalse(HDFS(_base_dir).exists(), "Base dir was not removed")
def should_create_directory_recursively(self):
_base_dir = os.path.join('/tmp', str(uuid.uuid4()))
_path = os.path.join(_base_dir, str(uuid.uuid4()), str(uuid.uuid4()))
_dir = HDFS(_path)
self.assertFalse(_dir.exists(), "Folder is already exists")
try:
_dir.create_directory(recursive=True)
self.assertTrue(_dir.exists(), "Folder was not created")
self.assertTrue(_dir.is_directory(), "New file should be a directory")
finally:
HDFS(_base_dir).delete_directory()
self.assertFalse(_dir.exists(), "File was not removed")
self.assertFalse(HDFS(_base_dir).exists(), "Base dir was not removed")
def test_get_modification_time(self):
now = datetime.now().strftime("%Y-%m-%d")
_dir = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
_file = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
try:
_dir.create_directory()
_file.create_file()
self.assertTrue(_dir.exists(), "Dir was not created")
self.assertTrue(_file.exists(), "File was not created")
self.assertEqual(now, _dir.modification_time().strftime("%Y-%m-%d"), "Error: dir modification time")
self.assertEqual(now, _file.modification_time().strftime("%Y-%m-%d"), "Error: File modification time")
finally:
_dir.delete_directory()
_file.delete()
def test_get_modification_time(self):
now = datetime.now().strftime("%Y-%m-%d")
_dir = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
_file = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
try:
_dir.create_directory()
_file.create_file()
self.assertTrue(_dir.exists(), "Dir was not created")
self.assertTrue(_file.exists(), "File was not created")
self.assertEqual(now, _dir.modification_time().strftime("%Y-%m-%d"), "Error: dir modification time")
self.assertEqual(now, _file.modification_time().strftime("%Y-%m-%d"), "Error: File modification time")
finally:
_dir.delete_directory()
_file.delete()
def test_mr_job_command_generation_with_arguments(self):
_job_name = "test_mr_job_%s" % uuid.uuid4()
_base_dir = HDFS(os.path.join("/tmp", _job_name))
_base_dir.create_directory()
try:
jar = os.path.join(os.path.dirname(__file__), 'resources', 'mapreduce', 'hadoop-mapreduce-examples.jar')
# configure job inputs
_job_input = HDFS(os.path.join(_base_dir.path, "input"))
_job_input.create_directory()
LocalFS(os.path.join(
os.path.dirname(__file__),
'resources',
'mapreduce', 'raw-data.txt')
).copy_to_hdfs(
_job_input.path
)
# configure job output
_job_output = HDFS(os.path.join(_base_dir.path, "output"))
if not os.path.exists(jar):
self.skipTest("'%s' not found" % jar)
job = MapReduce.prepare_mapreduce_job(jar=jar,
main_class="wordcount",
name=_job_name) \
.with_config_option("split.by", "'\\t'") \
.with_number_of_reducers(3) \
.with_arguments(
_job_input.path,
_job_output.path
)
_command_submission_result = job.run()
_command_submission_result.if_failed_raise(AssertionError("Cannot run MR job"))
_job_status = job.status()
self.assertTrue(_job_status is not None and _job_status.is_succeeded(), "MR job Failed")
self.assertTrue(_job_output.exists(), "Error: empty job output")
# check counters
self.assertEqual(6, _job_status.counter(group='File System Counters',
counter='HDFS: Number of write operations'))
self.assertEqual(1, _job_status.counter(group='Job Counters', counter='Launched map tasks'))
self.assertEqual(3, _job_status.counter(group='Job Counters', counter='Launched reduce tasks'))
self.assertEqual(2168, _job_status.counter(group='File Input Format Counters', counter='Bytes Read'))
finally:
_base_dir.delete_directory()
def test_streaming_job_with_multiple_inputs(self):
_job_basedir = HDFS(os.path.join("/tmp", str(uuid.uuid4())))
try:
job = self._template_streaming_job_(base_dir=_job_basedir.path)
_additional_datasource = HDFS(os.path.join(_job_basedir.path, "input2"))
_additional_datasource.create_directory()
LocalFS(os.path.join(os.path.dirname(__file__), 'resources',
'mapreduce', 'raw-data.txt')
).copy_to_hdfs(
_additional_datasource.path)
job.take(_additional_datasource.path)
command_result = job.run()
command_result.if_failed_raise(AssertionError("test_streaming_job_with_multiple_inputs test failed"))
_job_status = job.status()
self.assertTrue(_job_status is not None and _job_status.is_succeeded())
# check counters
self.assertEqual(740, _job_status.counter(group='Map-Reduce Framework', counter='Reduce input records'),
"counters['Map-Reduce Framework']['Reduce input records']")
finally:
_job_basedir.delete_directory()
def _template_streaming_job_(self, base_dir="/tmp", map_only_job=False):
if not os.path.exists(HADOOP_STREAMING_JAR):
self.skip("Cannot allocate %s" % HADOOP_STREAMING_JAR)
_hdfs_basdir = HDFS(base_dir)
if not _hdfs_basdir.exists():
_hdfs_basdir.create_directory()
_job_input = HDFS(os.path.join(_hdfs_basdir.path, "input"))
_job_input.create_directory()
_job_output = HDFS(os.path.join(_hdfs_basdir.path, "output"))
home = os.path.dirname(__file__)
_mapper = os.path.join(home, 'resources', 'mapreduce', 'mapper.py')
_reducer = os.path.join(home, 'resources', 'mapreduce', 'reducer.py')
LocalFS(
os.path.join(os.path.dirname(__file__), 'resources', 'mapreduce', 'raw-data.txt')
).copy_to_hdfs(
_job_input.path
)
return MapReduce.prepare_streaming_job(name="test-mr-streaming-job{}".format(str(uuid.uuid4())), jar=HADOOP_STREAMING_JAR) \
.take(_job_input.path) \
.process_with(mapper=_mapper, reducer=None if map_only_job else _reducer) \
.save(_job_output.path)
# may be used to endorse or promote products derived from this software without
# specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# See the NOTICE file and the LICENSE file distributed with this work
# for additional information regarding copyright ownership and licensing.
#
import os
from merlin.fs.hdfs import HDFS
from merlin.fs.localfs import LocalFS
if __name__ == "__main__":
_basedir = HDFS(os.path.join('/tmp', 'scd.active'))
_basedir.create_directory()
_scd_active_snapshot = LocalFS(
os.path.join(os.path.dirname(__file__), 'resources', 'scd.active.csv'))
_scd_active_snapshot.copy_to_hdfs(_basedir.path)
# specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# See the NOTICE file and the LICENSE file distributed with this work
# for additional information regarding copyright ownership and licensing.
#
import os
from merlin.fs.hdfs import HDFS
from merlin.fs.localfs import LocalFS
if __name__ == "__main__":
_basedir = HDFS(os.path.join('/tmp', 'scd.active'))
_basedir.create_directory()
_scd_active_snapshot = LocalFS(os.path.join(os.path.dirname(__file__), 'resources', 'scd.active.csv'))
_scd_active_snapshot.copy_to_hdfs(_basedir.path)
file.close()
if __name__ == '__main__':
log = get_logger("SCD")
# Prepare paths
_pig_script = os.path.join(os.path.dirname(__file__), 'scd_processing.pig')
_scd_active_snapshot = '/tmp/scd.active/scd.active.csv'
_scd_updates = os.path.join(os.path.dirname(__file__), 'resources', 'scd.update.csv')
_hdfs_job_output = '/tmp/scd.updated'
_local_folder_to_monitor = LocalFS(os.path.join(os.path.dirname(__file__), 'resources'))
_hdfs_basedir = HDFS('/tmp/scd.active')
_hdfs_tmpdir = HDFS('/tmp/scd.tmp')
_hdfs_tmpdir.create_directory()
if _scd_updates and LocalFS(_scd_updates).exists():
# Checks if file with last failed step is exists
# and reads this step
step = 'Copying scd updates to raw area on HDFS'
if os.path.isfile('resources/step'):
file = open('resources/step', 'r')
step = file.read()
file.close()
flow = FlowRegistry.flow('Flow')
# Runs flow
_context = flow.run(action=step,
if __name__ == '__main__':
log = get_logger("SCD")
# Prepare paths
_pig_script = os.path.join(os.path.dirname(__file__), 'scd_processing.pig')
_scd_active_snapshot = '/tmp/scd.active/scd.active.csv'
_scd_updates = os.path.join(os.path.dirname(__file__), 'resources',
'scd.update.csv')
_hdfs_job_output = '/tmp/scd.updated'
_local_folder_to_monitor = LocalFS(
os.path.join(os.path.dirname(__file__), 'resources'))
_hdfs_basedir = HDFS('/tmp/scd.active')
_hdfs_tmpdir = HDFS('/tmp/scd.tmp')
_hdfs_tmpdir.create_directory()
if _scd_updates and LocalFS(_scd_updates).exists():
# Checks if file with last failed step is exists
# and reads this step
step = 'Copying scd updates to raw area on HDFS'
if os.path.isfile('resources/step'):
file = open('resources/step', 'r')
step = file.read()
file.close()
flow = FlowRegistry.flow('Flow')
# Runs flow
_context = flow.run(