dpenfoldbrown

def date_to_datetime(date_str, fstrs=None):
    """
    Given date string, return datetime object
    2015 format: 2/8/13 0:00
    2016 format: 02/08/2013 00:00
                 02/12/2013 20:05
    """
    if fstrs is None:
        fstrs = ["%m/%d/%y %H:%M", "%m/%d/%Y %H:%M"]
    for f in fstrs:

dpenfoldbrown

import argparse

parser = argparse.ArgumentParser(description="Sample usage of argparse")

# Add argument. Do not use all elements as shown. Note: for flags, action="store_true" (?syntax?) is better.
parser.add_argument("-a", "--first", action="store", type=int, dest="first_arg", required=True, default=23, 
	help="An example first argument. Generally use only one of required or default. Leave out type for default (str).")

# Defaults to parsing sys.argv list, but can pass a list as well
args = parser.parse_args()

dpenfoldbrown

# Do this to update URL documents (assuming you already have url_dic and all other lists and dicts)

for user in collection.find():
    for tweet in user['tweets']:        
        urls = tweet['urls']
        for url in urls:
            
            url_string = url['url']             # DO NOT FORGET THIS. url is the url OBJECT, not the url string
            
            if url_string in short_urls:

dpenfoldbrown

# List of urls (pretend like it's populated)
urls = []

# Patterns to match in urls (note in some cases including the .org or .com to avoid matching common words or letters
# (eg for npr or slate or today)
# Add whatever other domains you want to match to the re OR (|) string
left_pattern = r"(?P<domain>nytimes|washingtonpost|npr.org|abcnews|nbcnews|huffingtonpost|slate.com|today.com)"
center_pattern = r"(?P<domain>cnn|bbc.co.uk|yahoo)"
right_pattern = r"(?P<domain>foxnews|washingtontimes|usnews|chicagotribune)"

dpenfoldbrown

d = { 'a':1, 'b':2, 'c':3, 'd':4 }
outhandle = open("dictfile.txt", 'w')
for (key, val) in d.items():
    outhandle.write("{0}\t{1}\n".format(key, val))
outhandle.close()

dpenfoldbrown

import pymongo

# Set up DB client, db, collection

date_tweet_count = {}
date_url_count = {}

for user in collection.find():
    for tweet in user['tweets']:
        datestr = tweet['created_at']

dpenfoldbrown

import pymongo

# Set up mongo DB here (get client, get database, get collection
client = MongoClient("smapp", 27011)
dbh = client['GunControl']
collection = dbh['GunTweetUsers_zephyr']

url_count = {}
tweet_count = {}

dpenfoldbrown

# Simple re examples in python

import re

files = ("re_pro_p00012.txt", "re_neg_p00014.txt", "lrr_p00014.txt")

category_pattern = r"(?P<category>[a-zA-Z]+)_.*"

for file in files:
    match = re.match(category_pattern, file)