python操作hdfs

import contextlib

import pyhdfs


class HdfsUtil(object):
    def __init__(self, hosts='namenode1:9870,namenode2:9870', user_name='hdfs'):
        self.hosts = hosts
        self.user_name = user_name
        self.fs = pyhdfs.HdfsClient(hosts=self.hosts, user_name=self.user_name)

    def listdir(self, dir_name):
        return self.fs.listdir(dir_name)

    def copy_from_local(self, local_file, hdfs_file):
        return self.fs.copy_from_local(local_file, hdfs_file)

    def copy_to_local(self, hdfs_file, local_file):
        return self.fs.copy_to_local(hdfs_file, local_file)

    def delete(self, hdfs_file, recursive=True):
        return self.fs.delete(hdfs_file, recursive=recursive)

    def exists(self, hdfs_file):
        return self.fs.exists(hdfs_file)

    def get_fs(self):
        return self.fs

    def create(self, hdfs_file):
        return self.fs.create(hdfs_file, ''.encode('utf8'))

    def read(self, hdfs_file):
        result = []
        with contextlib.closing(self.fs.open(hdfs_file)) as f:
            line = f.readline()
            while line:
                result.append(str(line, encoding='utf8'))
                line = f.readline()
        return ''.join(result)

    def write(self, hdfs_file, data: bytes):
        if not self.exists(hdfs_file):
            self.create(hdfs_file)
        self.fs.append(hdfs_file, data)
原文地址:https://www.cnblogs.com/wangbin2188/p/14591230.html