Florents-Tselai

#!/usr/bin/env bash

for url in https://www.nytimes.com/section/todayspaper https://www.washingtonpost.com/todays_paper/updates https://economist.com https://www.wsj.com/news/types/bookshelf https://www.nytimes.com/section/books https://www.economist.com/books-and-arts/ https://newyorker.com https://theatlantic.com https://www.prospectmagazine.co.uk/category/arts-and-books https://www.washingtonpost/entertainment/books https://news.ycombinator.com/best; do
    curl -L -H "Content-Type: application/json; charset=UTF-8" -H "X-Accept: application/json" \
-d "{\"url\": \"$url\", \"consumer_key\": \"$POCKET_CONSUMER_KEY\",\"access_token\": \"$POCKET_ACCESS_TOKEN\"}" \
-X POST "https://getpocket.com/v3/add"
done


# WSJ requires special care

Florents-Tselai

import tweepy

from datetime import datetime

auth = tweepy.OAuthHandler('CONSUMER_KEY', 'CONSUMER_SECRET')
auth.set_access_token('ACCESS_TOKEN', 'ACCESS_SECRET')

api = tweepy.API(auth, wait_on_rate_limit=True)

def parse_ts(ts):

Florents-Tselai

# Fundamentals
ipython
jupyter
pip
virtualenv

# Data/Scientific Related packages
pandas
numpy
scikit-learn

Florents-Tselai

# Adapted from http://stackoverflow.com/a/3452888
pip freeze --local | grep -v '^\-e' | cut -d = -f 1 | parallel pip install -U {} 

Florents-Tselai

#!/usr/bin/env python
import urllib2
import re
import sys
from itertools import imap, ifilter

if __name__ == "__main__":
    in_url = sys.argv[1]
    current_dir = "/".join(in_url.split('/')[:-1])
    all_urls = re.findall(r'href=[\'"]?([^\'" >]+)', urllib2.urlopen(in_url).read())