ajosanchez

from multiprocessing import dummy
import requests
import pymongo
import time

# install mongo db via url below
# https://docs.mongodb.com/manual/tutorial/install-mongodb-on-ubuntu/
mc = pymongo.MongoClient()
db = mc['db_name_here']
items = db['collection_name_here']

ajosanchez

import json

articles = []
with open('people.json', 'rb') as f:
    for item in f:
        articles.append(json.loads(item))

ajosanchez

from bs4 import BeautifulSoup as bs
import spacy

def hash_text(text, digits=8):
    return hash(text) % (10 ** digits)

def make_edge_dict(unique_edges):
	edge_dict = {}
	for edge in unique_edges:
		try:

ajosanchez

import csv
import json

def make_adjacency_list(adj_list):
    graph_list = []
    people = {}
    for person in adj_list:
        for k,v in person[1].items():
            person_id = person[0].keys()[0]
            graph_list.append((person_id, k))

ajosanchez

FROM ubuntu:18.04

RUN apt-get update

RUN apt-get install wget -y
RUN wget https://packages.microsoft.com/config/debian/10/packages-microsoft-prod.deb -O packages-microsoft-prod.deb && \
    dpkg -i packages-microsoft-prod.deb

RUN apt-get install -y apt-transport-https && \
    apt-get update && \

ajosanchez

# start the mongodb with the CLI command "mongod"
# you may get an error that the data path folder doesn't exist
# use mongod --dbpath /usr/local/mongodb-data to set the new directory for your data. dir must exist before command can be run

from pymongo import MongoClient
client = MongoClient()

db = client['test-database']

# create a collection called posts