本文整理汇总了Python中hdfs3.HDFileSystem类的典型用法代码示例。如果您正苦于以下问题:Python HDFileSystem类的具体用法?Python HDFileSystem怎么用?Python HDFileSystem使用的例子?那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。
在下文中一共展示了HDFileSystem类的15个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: read_block_from_hdfs
def read_block_from_hdfs(filename, offset, length, host=None, port=None,
delimiter=None):
from locket import lock_file
with lock_file('.lock'):
hdfs = HDFileSystem(host=host, port=port)
bytes = hdfs.read_block(filename, offset, length, delimiter)
return bytes
示例2: test_connection_error
def test_connection_error():
with pytest.raises(RuntimeError) as ctx:
hdfs = HDFileSystem(host='localhost', port=9999, connect=False)
hdfs.connect()
# error message is long and with java exceptions, so here we just check
# that important part of error is present
msg = 'Caused by: HdfsNetworkConnectException: Connect to "localhost:9999"'
assert msg in str(ctx.value)
示例3: read_block_from_hdfs
def read_block_from_hdfs(host, port, filename, offset, length, delimiter):
from hdfs3 import HDFileSystem
if sys.version_info[0] == 2:
from locket import lock_file
with lock_file('.lock'):
hdfs = HDFileSystem(host=host, port=port)
bytes = hdfs.read_block(filename, offset, length, delimiter)
else:
hdfs = HDFileSystem(host=host, port=port)
bytes = hdfs.read_block(filename, offset, length, delimiter)
return bytes
示例4: hdfs
def hdfs():
hdfs = HDFileSystem(host='localhost', port=8020)
if hdfs.exists('/tmp/test'):
hdfs.rm('/tmp/test')
hdfs.mkdir('/tmp/test')
yield hdfs
if hdfs.exists('/tmp/test'):
hdfs.rm('/tmp/test')
示例5: make_hdfs
def make_hdfs():
hdfs = HDFileSystem(host='localhost', port=8020)
if hdfs.exists('/tmp/test'):
hdfs.rm('/tmp/test')
hdfs.mkdir('/tmp/test')
try:
yield hdfs
finally:
if hdfs.exists('/tmp/test'):
hdfs.rm('/tmp/test')
示例6: __init__
def __init__(self):
"""Initializes a connection to HDFS.
Connection configuration is done using :doc:`hdfs`.
"""
super(HadoopFileSystem, self).__init__()
self._hdfs_client = HDFileSystem()
示例7: make_hdfs
def make_hdfs():
from hdfs3 import HDFileSystem
# from .hdfs import DaskHDFileSystem
basedir = '/tmp/test-distributed'
hdfs = HDFileSystem(host='localhost', port=8020)
if hdfs.exists(basedir):
hdfs.rm(basedir)
hdfs.mkdir(basedir)
try:
yield hdfs, basedir
finally:
if hdfs.exists(basedir):
hdfs.rm(basedir)
示例8: get_block_locations
def get_block_locations(self, paths):
offsets = []
lengths = []
machines = []
for path in paths:
if path.startswith('hdfs://'):
path = path[len('hdfs://'):]
out = HDFileSystem.get_block_locations(self, path)
offsets.append([o['offset'] for o in out])
lengths.append([o['length'] for o in out])
machines.append([o['hosts'] for o in out])
return offsets, lengths, machines
示例9: hdfs_open_file
def hdfs_open_file(path, auth):
hdfs = HDFileSystem(**auth)
return hdfs.open(path, mode='rb')
示例10: HDFileSystem
from hdfs3 import HDFileSystem
import os
os.environp['HADOOP_USER_NAME']=hadoop
hdfs = HDFileSystem(host='trevally.amer.nevint.com', port=9000)
print hdfs.ls('/user/hadoop')
示例11: __init__
def __init__(self, **kwargs):
kwargs2 = {k: v for k, v in kwargs.items()
if k in ['host', 'port', 'user', 'ticket_cache',
'token', 'pars']}
HDFileSystem.__init__(self, connect=True, **kwargs2)
示例12: import
from tornado import gen
from dask.imperative import Value
from distributed.utils_test import gen_cluster, cluster, loop, make_hdfs
from distributed.utils import get_ip
from distributed.hdfs import (read_bytes, get_block_locations, write_bytes,
_read_csv, read_csv)
from distributed import Executor
from distributed.executor import _wait, Future
pytest.importorskip('hdfs3')
from hdfs3 import HDFileSystem
try:
hdfs = HDFileSystem(host='localhost', port=8020)
hdfs.df()
del hdfs
except:
pytestmark = pytest.mark.skipif('True')
ip = get_ip()
def test_get_block_locations():
with make_hdfs() as hdfs:
data = b'a' * int(1e8) # todo: reduce block size to speed up test
fn_1 = '/tmp/test/file1'
fn_2 = '/tmp/test/file2'
示例13: open
def open(self, path, mode='rb', **kwargs):
if path.startswith('hdfs://'):
path = path[len('hdfs://'):]
return HDFileSystem.open(self, path, mode, **kwargs)
示例14: hdfs_open_file
def hdfs_open_file(path, auth):
from hdfs3 import HDFileSystem
hdfs = HDFileSystem(**auth)
return hdfs.open(path, mode='rb')
示例15: open
def open(self, path, mode='rb'):
mode = mode.rstrip('b')
return open(path, mode)
def put(self, src, dst):
return shutil.copy(src, dst)
if __name__ == "__main__":
# load the hdfs node info
f = open('hdfs.yml', 'r')
data = yaml.load(f)
f.close()
hdfs_nn = data['hdfs_nn']
hdfs = HDFileSystem(host=hdfs_nn, port=data['hdfs_port'])
tfs = TransparentFileSystem(hdfs)
print hdfs.exists('/tmp')
# print hdfs.hoge('/tmp')
print tfs.exists('/tmp')
# print tfs.hoge('/tmp')
# tfs_local = TransparentFileSystem()
# print tfs_local.glob('/var/tmp')
print 'test'
print tfs.glob('/tmp')
# tfs.hoge()
tfs_local = TransparentFileSystem()
# print tfs_local.glob('/home/vagrant/work/data/*')