In [48]:

    

import datetime as dt
from steemdata import SteemData

from funcy.colls import pluck
from funcy.seqs import mapcat
from funcy.flow import silent


    

In [2]:

    

from steem.post import Post


    

In [3]:

    

%load_ext autoreload
%autoreload 2


    

In [4]:

    

db = SteemData().db


    

In [31]:

    

time_constraints = {
    '$gte': dt.datetime.now() - dt.timedelta(days=7),
}
conditions = {
    'created': time_constraints,
}
projection = {
    '_id': 0,
    'identifier': 1,
    'abs_rshares': 1,
    'net_rshares': 1,
    'vote_rshares': 1,
    'children_abs_rshares': 1,
}
posts = list(db['Posts'].find(conditions, projection=projection))


    

In [32]:

    

## get the comments


    

In [33]:

    

identifiers = list(pluck('identifier', posts))


    

In [34]:

    

len(identifiers)


    

    
Out[34]:





10145

In [13]:

    

def get_comments(identifier):
    try:
        p = Post(identifier)
    except:
        return []
    return Post.get_all_replies(p)


    

In [51]:

    

posts = map(silent(Post), identifiers)
# remove empty ones
posts = list(filter(bool, posts))


    

In [14]:

    

comments = mapcat(get_comments, identifiers)


    

In [18]:

    

## tally up rshares


    

In [58]:

    

def vote_rshares_sum(post):
    return sum(map(int, pluck('rshares', post.active_votes)))


    

In [59]:

    

def rshares_sum(posts):
    return sum(map(vote_rshares_sum, posts))


    

In [60]:

    

post_rshares = rshares_sum(posts)


    

In [70]:

    

comment_rshares = rshares_sum(comments)


    

In [ ]:

    

comment_rshares / post_rshares * 100