CODEWITHSUNDEEP
pythonoauthtumblr
JustaCookie
JustaCookie

Reputation: 181

Python 3 with Requests trying to use Tumblr API, I get error 401?

I have Python 3 and I am trying to post to Tumblr via API link to API documentation. I keep getting an error 401 despite feeling like I am doing everything correctly. There is an official API client in Python 2 but that feels a bit hard to follow and all other mentions of it seem to be in PHP or Java. I also am not sure of the format to post in after the 401 error because the documentation doesn't give any explicit examples other than /post. My code:

import requests
from requests_oauthlib import OAuth1
#variables for later
client_key=""
client_secret=""
oauth_token=""
oauth_token_secret=""

#gets the values for the variables
with open("API.txt", 'r') as readAPI:
    readAPI.readline()
    client_key=readAPI.readline()[23:]
    client_secret=readAPI.readline()[23:]
    oauth_token=readAPI.readline()[23:]
    oauth_token_secret=readAPI.readline()[23:]
readAPI.close()

#prints them to double check they are being read correctly
print(client_key)
print(client_secret)
print(oauth_token)
print(oauth_token_secret)

#sets oauth for the connection
oauth = OAuth1(client_key,
               client_secret,
               oauth_token,
               oauth_token_secret)

#check post that should return various blog stats
r = requests.get("http://api.tumblr.com/v2/user/info" ,auth=oauth)

print(r)

I am 100% sure I am getting the client key, the client secret, oauth token and oauth token secret correct. I have double checked, the oauth tokens are both put in the file that is being read manually and they are printed before the connection attempt. I am 100% sure it is correct. I'm wondering if Tumblr's API is broken?

Edit: This is with print(repr())

'client_key'
'client_secret'
'oauth_token'
'oauth_token_secret'
{"meta":{"status":401,"msg":"Not Authorized"},"response":[]}

This is what happens after trying a new code and with Steven's method with JSON.

b'{"meta":{"status":401,"msg":"Not Authorized"},"response":[]}'

Upvotes: 0

Views: 989

Answers (1)

Steven Kryskalla
Steven Kryskalla

Reputation: 14649

Instead of doing this:

print(client_key)

What is the output of this?

print(repr(client_key))

You're using readline, which includes a newline character at the end of each line:

$ cat foo.txt
key
secret
blabla

$ python3.4
>>> f = open("foo.txt")
>>> print(repr(f.readline()))
'key\n'
>>> print(repr(f.readline()))
'secret\n'
>>> print(repr(f.readline()))
'blabla\n'

Have you tried stripping the newline character off of each line?

Edit: Updating my post based on @user2853325's comments. Your code works for me under Python 3.4, requests==2.5.2, and requests-oauthlib==0.4.2.

API.json (redacted the keys/secrets):

{
    "client_key": "XXXXXXXXXXXXXXXXXXXXdG7zXIMcDidwQ5pMHuQTbxyhNINrCE",
    "client_secret": "XXXXXXXXXXXXXXXXXXXX72A5HQO1axydP5nlOWCTQx4ECfXfyX",
    "oauth_token": "XXXXXXXXXXXXXXXXXXXX8WAnqMBWaAdnGhnc4gWhJ4j6cufK1W",
    "oauth_token_secret": "XXXXXXXXXXXXXXXXXXXX8Kf82k65JzIcMU7QUp54ssPEzJd7my"
}

tumblr.py:

import json

import requests
from requests_oauthlib import OAuth1

#gets the values for the variables
with open("API.json") as f:
    credentials = json.load(f)

#prints them to double check they are being read correctly
print(credentials)

#sets oauth for the connection
oauth = OAuth1(
    credentials['client_key'],
    credentials['client_secret'],
    credentials['oauth_token'],
    credentials['oauth_token_secret']
)

#check post that should return various blog stats
r = requests.get("http://api.tumblr.com/v2/user/info", auth=oauth)
print(r)
print(r.content)

Output (redacted the oauth stuff):

$ bin/python tumblr.py
{'oauth_token_secret': 'XXXXXXXXXXXXXXXXXXXX8Kf82k65JzIcMU7QUp54ssPEzJd7my', 'client_secret': 'XXXXXXXXXXXXXXXXXXXX72A5HQO1axydP5nlOWCTQx4ECfXfyX', 'client_key': 'XXXXXXXXXXXXXXXXXXXXdG7zXIMcDidwQ5pMHuQTbxyhNINrCE', 'oauth_token': 'XXXXXXXXXXXXXXXXXXXX8WAnqMBWaAdnGhnc4gWhJ4j6cufK1W'}
<Response [200]>
b'{"meta":{"status":200,"msg":"OK"},"response":{"user":{"name":"lost-theory","likes":0,"following":2,"default_post_format":"html","blogs":[{"title":"Untitled","name":"lost-theory","posts":0,"url":"http:\\/\\/lost-theory.tumblr.com\\/","updated":0,"description":"","is_nsfw":false,"ask":false,"ask_page_title":"Ask me anything","ask_anon":false,"followed":false,"can_send_fan_mail":true,"share_likes":true,"likes":0,"twitter_enabled":false,"twitter_send":false,"facebook_opengraph_enabled":"N","tweet":"N","facebook":"N","followers":0,"primary":true,"admin":true,"messages":0,"queue":0,"drafts":0,"type":"public"}]}}}'

So now that I've tested your code out for myself:

  • How did you get the oauth_token and oauth_token_secret? I got mine by clicking "Explore API" on the Applications developer page.
  • You don't need to call readAPI.close(), as the with block automatically closes the file for you (see the official docs).
  • I used JSON for storing and reading the credentials, that way I'm 100% sure that I'm getting the correct strings. The code is cleaner too. I'm still suspicious about the way you're reading lines from the file and slicing them.
  • Try printing r.content in your code the same way I am. Does it give you a more descriptive error message than "401 Unauthorized"?

Upvotes: 1

Related Questions