如何使用python在hadoop文件系统中创建json文件

Question

 os.system('echo "%s" | hadoop fs -put -f - /app/hdp/logs/json/a.json' %(json_string))

json_string=json.dumps({"a":"b"})

The file got created in HDFS but with wrong json format. The format it stored was {a:b} with no doubles quotes which is not a proper json format.

What is wrong in this approach?

Answer 1

Try using the following:

import subprocess, json

json_string=json.dumps({"a":"b"})

proc = subprocess.Popen('echo "{0}" | hadoop fs -put -f - /app/hdp/logs/json/a.json'.format(json_string), shell=True)

The string should be formatted as "{"a":"b"}"

Answer 2

import subprocess, json
json_string=json.dumps({"a":"b"})
#json_string=json_string.replace('"','\"') try escaping quotes too
proc = subprocess.run('echo {0} | hadoop fs -put -f - /app/hdp/logs/json/a.json'.format(json_string), shell=True)

Answer 3

You will need to include escape characters

echo "{"a":"b"}" in shell terminal

Output: {a:b}