Upload attachment to Confluence using Nifi ExecuteScript in Python

Question

I'm trying to upload a PDF file to Confluence using Nifi's ExecuteScript processor. I can upload the file successfully, but when I download and open it, it's BLANK. There must be something wrong with my conversion. Can anyone please help check?

So this is how I do it:

1. download the PDF file from an internal API
2. ExecuteScript Groovy - to convert the flowfile content to attribute

import org.apache.commons.io.IOUtils
import java.nio.charset.StandardCharsets

flowFile = session.get()
if(!flowFile)return
def text = ''
session.read(flowFile, {inputStream ->
  text = IOUtils.toString(inputStream, StandardCharsets.UTF_8)
} as InputStreamCallback)

flowFile = session?.putAttribute(flowFile, "file_content", text)
session.transfer(flowFile, /*ExecuteScript.*/ REL_SUCCESS)

3. ExecuteScript Python - to upload PDF file to Confluence

Here's my code for #3. I think something's wrong here -->

import json
import requests
from requests_toolbelt.multipart.encoder import MultipartEncoder
from org.apache.nifi.processor.io import OutputStreamCallback

class OutputWrite(OutputStreamCallback):
  def __init__(self, obj):
  self.obj = obj
  def process(self, outputStream):
     outputStream.write(bytearray(json.dumps(self.obj).encode('utf-8')))

flowFile = session.get()
if (flowFile != None):
  url = 'https://myconfluence.com/rest/api/content/12345/child/attachment'
  auth = 'myauthorization'
  file_name = 'mypdf.pdf'
  file_content = flowFile.getAttribute('file_content')

  s = requests.Session()

  m = MultipartEncoder(fields={'file': (file_name, file_content, 'application/pdf')})
  headers = {"X-Atlassian-Token":"nocheck", "Authorization":auth, "Content-Type":m.content_type}

  r = s.post(url, data=m, headers=headers, verify=False)

  session.write(flowFile, OutputWrite(json.loads(r.text)))
  session.transfer(flowFile, REL_SUCCESS)
  session.commit()

---

UPDATE 06/28/2019

I decided to follow Peter's advice and merge codes 1 and 2. It's still not working. Before, the PDF file is 2MB, but it's BLANK. Now, its size is 0KB. Any help would be greatly appreciated!

import json
import requests
from requests_toolbelt.multipart.encoder import MultipartEncoder
from org.apache.nifi.processor.io import OutputStreamCallback
from org.apache.commons.io import IOUtils
from java.nio.charset import StandardCharsets
from org.apache.nifi.processor.io import InputStreamCallback

class PyInputStreamCallback(InputStreamCallback):
    def __init__(self):
        pass
    def process(self, inputStream):
        text = IOUtils.toString(inputStream, StandardCharsets.UTF_8)

class OutputWrite(OutputStreamCallback):
    def __init__(self, obj):
        self.obj = obj
    def process(self, outputStream):
        outputStream.write(bytearray(json.dumps(self.obj).encode('utf-8')))

text = ''
flowFile = session.get()
if(flowFile != None):
    session.read(flowFile, PyInputStreamCallback())
    confluence_attachment_api = flowFile.getAttribute('confluence_attachment_api')
    confluence_authorization = flowFile.getAttribute('confluence_authorization')
    file_name = flowFile.getAttribute('file_name')

    s = requests.Session()
    m = MultipartEncoder(fields={'file': (file_name, text, 'application/pdf')})
    headers = {"X-Atlassian-Token":"nocheck", "Authorization":confluence_authorization, "Content-Type":m.content_type}
    r = s.post(confluence_attachment_api, data=m, headers=headers, verify=False)

    session.write(flowFile, OutputWrite(json.loads(r.text)))
    session.transfer(flowFile, REL_SUCCESS)
    session.commit()

Answer 1

It doesn't look like you are actually sending the FlowFile contents. Instead, you are sending an attribute named file_content as the file contents, which probably isn't what you intended

You will need to do a session.read to get the file stream. The below code doesn't work as is, but shows how you can get access to the stream.

class PyInputStreamCallback(InputStreamCallback):
  def __init__(self):
        pass
  def process(self, inputStream):
    m = MultipartEncoder(fields={'file': (file_name, inputStream, 'application/pdf')})

session.read(flowFile, PyInputStreamCallback())

Ref: https://community.hortonworks.com/articles/75545/executescript-cookbook-part-2.html