def build_wordcount(db,collection=None): db = pymongo.Connection()[db] word_counts = pymongo.Connection()['word_counts'] min_time, max_time = v.get_time_bounds(collection = collection) start = min_time end = min_time + timedelta(minutes=2) while time.mktime(end.timetuple()) <= time.mktime(max_time.timetuple()): results = v.get_from_db(collection,start,end) for i1, result in enumerate(results): for i2, x in enumerate(result): results[i1][i2] = (x[0].decode('utf-8','replace'), x[1]) if word_counts[collection].insert({'timestamp':start, 'results':results}): print 'O DANG' start = end if time.mktime(end.timetuple()) == time.mktime(max_time.timetuple()): break end = end + timedelta(minutes=2) if time.mktime(end.timetuple()) > time.mktime(max_time.timetuple()): end = max_time
def build_choosetwo(collection=None): db = pymongo.Connection()['word_counts'] word_counts = pymongo.Connection()['word_counts'] min_time, max_time = v.get_time_bounds(collection = collection) start = min_time while start < max_time: timestamps = [start] data = [Counter(),Counter(),Counter(),Counter()] while max_time not in timestamps: data = aggregate_and_save(db, collection, timestamps, data) new_stamp = timestamps[ len(timestamps) - 1 ] + timedelta( minutes=2 ) if new_stamp > max_time: break timestamps.append( new_stamp ) start += timedelta( minutes=2 )