Important:

I'm just starting to learn Python and I'm not a developer.

Use the code below entirely at your own risk.

---

I'm slowly learning how to code with Python 3.
I took this bot apart and put it back together and - for the most part - understand how it works now:

https://steemit.com/socialist-bot/@fyrstikken/the-anonymous-winfrey-bot-upvotes-for-everyone-download-here-easy-steps-for-n00bs

It's very motivating to see your code doing something useful instantly.
Coding Python with Piston for curation, can be very rewarding; Mentally and monetarily.

As I learn new stuff from programming tutorials, I try to use them in my bot.

I'm sure I made some mistakes, but this is mainly an exercise for me.

It will need 4 files in the same directory to work:

accounts.txt	wif.txt	authors.txt	upvote_history.txt
account1	5wifkeyreallylong	goodauthor1
account2	5wifkeyevenlonger	goodauthor2

votebot.py:

import threading
import random
import time
import csv
from time import localtime
from steem.steem import Steem
from steem.steem import BroadcastingError

vote_delay = 1740

def feed():
    authors= list_load('authors.txt')
    upvote_history= list_load('upvotehistory.txt')
    steem = Steem(wif=posting_key)
    for comment in steem.stream_comments():
        if comment.author in authors:
            if len(comment.title) > 0:
                if comment.identifier in upvote_history: 
                    continue                              
                upvote_history.append(comment.identifier)
                list_save('upvotehistory.txt', upvote_history)
                print(timestamp_builder(), comment.identifier, \
                    ": thread started, vote in ", vote_delay, " seconds ...")
                workerThread = threading.Thread(name=comment.identifier, \
                    target=worker, args=(comment,))
                workerThread.start()

def worker(worker_comment):
    time.sleep(vote_delay)
    try:
      for (k,v) in enumerate(account):
        print(timestamp_builder(), worker_comment.identifier, " <--- voted ",v)
        worker_steem = Steem(wif=posting_key[k])
        upvote_comment = worker_steem.get_content(worker_comment.identifier)
        upvote_comment.vote(100, v)
    except BroadcastingError as e:
      print(str(E)) 

def list_save(listfile,  listvar):
    with open(listfile, 'w') as writestuff:
        writestuff.write('\n'.join(listvar))     
        writestuff.write('\n')

def list_load(listfile):
    with open(listfile, 'r') as readstuff:
        listvar= []
        reader= csv.reader(readstuff)
        for rows in reader:
            v= rows[0]
            listvar.append(v)   
    return listvar

def timestamp_builder(): 
    lt = localtime()
    timestamp = time.strftime("%d/%m/%Y-%H:%M:%S ", lt)
    return timestamp

account= list_load('accounts.txt')
posting_key= list_load('wif.txt')

if __name__ == "__main__":
    while True:
        try:
            feed()
        except (KeyboardInterrupt, SystemExit):
            print("Quit")
            break
        except Exception as e:
            print("Exception ... Restart")
            traceback.print_exc()

Here is an example of the printout:

10/01/2017-19:18:35  @pgarcgo/proyecto-cervantes-compensacion-y-reconocimiento-para-escritores-de-habla-hispana-vol-106 : thread started, vote in  1740  seconds ...
10/01/2017-19:47:35  @pgarcgo/proyecto-cervantes-compensacion-y-reconocimiento-para-escritores-de-habla-hispana-vol-106  <--- voted  votebot
10/01/2017-19:47:38  @pgarcgo/proyecto-cervantes-compensacion-y-reconocimiento-para-escritores-de-habla-hispana-vol-106  <--- voted  deutschbot

Additions:

• input/output functions to store accounts, wifs, authors, upvote_history in txt files
• timestamp function
• better handling for multiple accounts

Also, I changed when upvote_history gets updated.
This way, the bot won't start mutiple threads if the author edits his post before the bot voted.

upvote_history gets stored in a file now, so when the bot crashes and restarts, it won't try to upvote a post again.
This is still far from ideal.

It's still very basic and I don't think it's a good approach to curation in general.
I did it for the exercise and learned a lot.

Ideally, I would need a steem.stream_votes() function, similar to steem.stream_comments()

Next steps will be databases and maybe how to use a proper config file.