# -*- coding: utf-8 -*-
from collections import Counter
from datetime import datetime, timedelta
import falcon
import ujson as json
from streamer import get_db, get_keywords
from hortiradar import tokenizeRawTweetText, admins, users
tweets = get_db().tweets
KEYWORDS = get_keywords()
time_format = "%Y-%m-%d-%H-%M-%S"
with open("data/stoplist-nl.txt") as f:
stop_words = [w.decode("utf-8").strip() for w in f.readlines()]
stop_words = {w: 1 for w in stop_words} # stop words to filter out in word cloud
def get_dates(req, resp, resource, params):
"""Parse the 'start' and 'end' datetime parameters."""
try:
start = req.get_param("start")
start = datetime.strptime(start, time_format) if start else datetime(2001, 1, 1)
end = req.get_param("end")
end = datetime.strptime(end, time_format) if end else datetime(3001, 1, 1)
params["start"] = start
params["end"] = end
except ValueError:
msg = "Invalid datetime format string, use: %s" % time_format
from streamer import get_db, get_keywords, find_keywords_and_groups
from hortiradar import tokenizeRawTweetText
tweets = get_db().tweets
keywords = get_keywords()
tw = tweets.find()
for t in tw:
tokens = tokenizeRawTweetText(t["tweet"]["text"])
kws, groups = find_keywords_and_groups(tokens, keywords)
tweets.update_one({"_id": t["_id"]}, {
"$set": {
"keywords": kws,
"groups": groups,
"num_keywords": len(kws)
}
})
