-
Notifications
You must be signed in to change notification settings - Fork 0
/
scrape.py
38 lines (28 loc) · 1.26 KB
/
scrape.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
import csv
import os
from datetime import datetime
from confidence import Configuration, load_name
from scraper import load_scraper
from scraper.utils import AuthorHandler, clean, read_yaml_utf8
def main():
"""
Main script for scraping websites defined in config/scrape.yaml.
Writes results to csv as they are retrieved.
"""
config = load_name('config/scrape')
os.makedirs("scratch", exist_ok=True)
filename = f"scratch/results_{datetime.now().strftime('%Y%m%d-%H%M%S')}.csv"
author_handler = AuthorHandler(authors=list(read_yaml_utf8('config/local.yaml')['authors']))
global_settings = {'author_handler': author_handler}
with open(filename, "w", encoding='utf-8', newline="") as csvfile:
writer = csv.writer(csvfile)
writer.writerow(["page_author_id", "text", "page_name", "url"])
for website in config.websites:
scraper = load_scraper(website.kind, **Configuration(global_settings, website.settings))
website_posts = [post for post in scraper.run()]
for post in website_posts:
writer.writerow(
[post.author_id, author_handler.replace_authors(clean(post.text)), post.name, post.url]
)
if __name__ == '__main__':
main()