to_pickle.py

from nltk.tokenize import word_tokenize
import pickle
import pprint
import json

"""

(heads, descs, keywords) = ([headline], [description], )

"""

DATA_FILE = 'data/signalmedia-1m.jsonl.test'
PICKLE_FILE = 'data/tokens.pkl'


def write_to_pickle(filename, data):   
    with open(filename, 'wb') as f:
        pickle.dump(data, f)


def read_from_pickle(filename):    
    with open(filename, 'r') as f:
        return pickle.load(f)


def get_json_data(filename):    
    heads = []
    descs = []
    HEADER = "title"
    DESCRIPTION = "content"

    with open(filename, 'r') as json_data:
        for json_object in json_data:
            heads.append(json.loads(json_object)[HEADER])
            descs.append(json.loads(json_object)[DESCRIPTION])
    return heads, descs, None


data = get_json_data(DATA_FILE)
write_to_pickle(PICKLE_FILE, data)
data = read_from_pickle(PICKLE_FILE)
pprint.pprint(data)