forked from tanner/qotnews
Compare commits
5 Commits
736cdc8576
...
9f4ff4acf0
Author | SHA1 | Date |
---|---|---|
Jason Schwarzenberger | 9f4ff4acf0 | 4 years ago |
Jason Schwarzenberger | db6aad84ec | 4 years ago |
Jason Schwarzenberger | 29f8a8b8cc | 4 years ago |
Jason | abf8589e02 | 4 years ago |
Jason | b759f46582 | 4 years ago |
6 changed files with 199 additions and 115 deletions
@ -0,0 +1,180 @@ |
||||
import logging |
||||
logging.basicConfig( |
||||
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s', |
||||
level=logging.DEBUG) |
||||
|
||||
if __name__ == '__main__': |
||||
import sys |
||||
sys.path.insert(0,'.') |
||||
|
||||
import requests |
||||
from datetime import datetime |
||||
from bs4 import BeautifulSoup |
||||
import extruct |
||||
|
||||
from utils import clean |
||||
|
||||
OUTLINE_API = 'https://api.outline.com/v3/parse_article' |
||||
USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:77.0) Gecko/20100101 Firefox/77.0' |
||||
|
||||
def unix(date_str): |
||||
date_tzfix = date_str |
||||
if ":" == date_tzfix[-3]: |
||||
date_tzfix = date_tzfix[:-3]+date_tzfix[-2:] |
||||
formats = ['%Y-%m-%dT%H:%M:%SZ', '%Y-%m-%dT%H:%M:%S%z', '%Y-%m-%dT%H:%M:%S.%fZ', '%Y-%m-%dT%H:%M:%S.%f%z'] |
||||
for f in formats: |
||||
try: |
||||
return int(datetime.strptime(date_str, f).timestamp()) |
||||
except: |
||||
pass |
||||
try: |
||||
return int(datetime.strptime(date_tzfix, f).timestamp()) |
||||
except: |
||||
pass |
||||
return 0 |
||||
|
||||
def xml(route, ref=None): |
||||
try: |
||||
headers = {'User-Agent': USER_AGENT, 'X-Forwarded-For': '66.249.66.1'} |
||||
r = requests.get(route(ref), headers=headers, timeout=5) |
||||
if r.status_code != 200: |
||||
raise Exception('Bad response code ' + str(r.status_code)) |
||||
return r.text |
||||
except KeyboardInterrupt: |
||||
raise |
||||
except BaseException as e: |
||||
logging.error('Problem hitting URL: {}'.format(str(e))) |
||||
return False |
||||
|
||||
def parse_extruct(s, data): |
||||
for rdfa in data['rdfa']: |
||||
for key, props in rdfa.items(): |
||||
if 'http://ogp.me/ns#title' in props: |
||||
for values in props['http://ogp.me/ns#title']: |
||||
s['title'] = values['@value'] |
||||
if 'http://ogp.me/ns/article#modified_time' in props: |
||||
for values in props['http://ogp.me/ns/article#modified_time']: |
||||
print(f"modified_time: {values['@value']}") |
||||
s['date'] = unix(values['@value']) |
||||
if 'http://ogp.me/ns/article#published_time' in props: |
||||
for values in props['http://ogp.me/ns/article#published_time']: |
||||
print(f"published_time: {values['@value']}") |
||||
s['date'] = unix(values['@value']) |
||||
|
||||
for og in data['opengraph']: |
||||
titles = list(filter(None, [value if 'og:title' in key else None for key, value in og['properties']])) |
||||
modified = list(filter(None, [value if 'article:modified_time' in key else None for key, value in og['properties']])) |
||||
published = list(filter(None, [value if 'article:published_time' in key else None for key, value in og['properties']])) |
||||
if len(modified): |
||||
s['date'] = unix(modified[0]) |
||||
if len(published): |
||||
s['date'] = unix(published[0]) |
||||
s['date'] = unix(published[0] or modified[0] or '') |
||||
if len(titles): |
||||
s['title'] = titles[0] |
||||
|
||||
for md in data['microdata']: |
||||
if md['type'] == 'https://schema.org/NewsArticle': |
||||
props = md['properties'] |
||||
s['title'] = props['headline'] |
||||
if props['dateModified']: |
||||
s['date'] = unix(props['dateModified']) |
||||
if props['datePublished']: |
||||
s['date'] = unix(props['datePublished']) |
||||
if 'author' in props and props['author']: |
||||
s['author'] = props['author']['properties']['name'] |
||||
|
||||
for ld in data['json-ld']: |
||||
if ld['@type'] == 'Article': |
||||
s['title'] = ld['headline'] |
||||
if ld['dateModified']: |
||||
s['date'] = unix(ld['dateModified']) |
||||
if ld['datePublished']: |
||||
s['date'] = unix(ld['datePublished']) |
||||
if 'author' in ld and ld['author']: |
||||
s['author'] = ld['author']['name'] |
||||
|
||||
return s |
||||
|
||||
class Sitemap: |
||||
def __init__(self, url): |
||||
self.sitemap_url = url |
||||
|
||||
def feed(self): |
||||
markup = xml(lambda x: self.sitemap_url) |
||||
if not markup: return [] |
||||
soup = BeautifulSoup(markup, features='lxml') |
||||
articles = soup.find('urlset').findAll('url') |
||||
articles = list(filter(None, [a if a.find('lastmod') is not None else None for a in articles])) |
||||
return [x.find('loc').text for x in articles] or [] |
||||
|
||||
def story(self, ref): |
||||
markup = xml(lambda x: ref) |
||||
if not markup: |
||||
return False |
||||
|
||||
s = {} |
||||
s['author_link'] = '' |
||||
s['score'] = 0 |
||||
s['comments'] = [] |
||||
s['num_comments'] = 0 |
||||
s['link'] = ref |
||||
s['url'] = ref |
||||
s['date'] = 0 |
||||
|
||||
data = extruct.extract(markup) |
||||
s = parse_extruct(s, data) |
||||
return s |
||||
|
||||
class Category: |
||||
def __init__(self, url): |
||||
self.category_url = url |
||||
self.base_url = '/'.join(url.split('/')[:3]) |
||||
|
||||
def feed(self): |
||||
markup = xml(lambda x: self.category_url) |
||||
if not markup: return [] |
||||
soup = BeautifulSoup(markup, features='html.parser') |
||||
links = soup.find_all('a', href=True) |
||||
links = [link.get('href') for link in links] |
||||
links = [f"{self.base_url}{link}" if link.startswith('/') else link for link in links] |
||||
links = list(filter(None, [link if link.startswith(self.category_url) else None for link in links])) |
||||
return links |
||||
|
||||
def story(self, ref): |
||||
markup = xml(lambda x: ref) |
||||
if not markup: |
||||
return False |
||||
|
||||
s = {} |
||||
s['author_link'] = '' |
||||
s['score'] = 0 |
||||
s['comments'] = [] |
||||
s['num_comments'] = 0 |
||||
s['link'] = ref |
||||
s['url'] = ref |
||||
s['date'] = 0 |
||||
|
||||
data = extruct.extract(markup) |
||||
s = parse_extruct(s, data) |
||||
return s |
||||
|
||||
# scratchpad so I can quickly develop the parser |
||||
if __name__ == '__main__': |
||||
print("Sitemap: Stuff") |
||||
site = Sitemap("https://www.stuff.co.nz/sitemap.xml") |
||||
posts = site.feed() |
||||
print(posts[:1]) |
||||
print(site.story(posts[0])) |
||||
|
||||
print("Sitemap: NZ Herald") |
||||
site = Sitemap("https://www.nzherald.co.nz/arcio/news-sitemap/") |
||||
posts = site.feed() |
||||
print(posts[:1]) |
||||
print(site.story(posts[0])) |
||||
|
||||
print("Category: RadioNZ Te Ao Māori") |
||||
site = Category("https://www.rnz.co.nz/news/te-manu-korihi/") |
||||
posts = site.feed() |
||||
print(posts[:1]) |
||||
print(site.story(posts[0])) |
@ -1,110 +0,0 @@ |
||||
import logging |
||||
logging.basicConfig( |
||||
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s', |
||||
level=logging.DEBUG) |
||||
|
||||
if __name__ == '__main__': |
||||
import sys |
||||
sys.path.insert(0,'.') |
||||
|
||||
import requests |
||||
from datetime import datetime |
||||
from bs4 import BeautifulSoup |
||||
|
||||
from utils import clean |
||||
|
||||
OUTLINE_API = 'https://api.outline.com/v3/parse_article' |
||||
USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:77.0) Gecko/20100101 Firefox/77.0' |
||||
|
||||
def unix(date_str): |
||||
return int(datetime.strptime(date_str, '%Y-%m-%dT%H:%M:%SZ').timestamp()) |
||||
|
||||
def xml(route, ref=None): |
||||
try: |
||||
headers = {'User-Agent': USER_AGENT, 'X-Forwarded-For': '66.249.66.1'} |
||||
r = requests.get(route(ref), headers=headers, timeout=5) |
||||
if r.status_code != 200: |
||||
raise Exception('Bad response code ' + str(r.status_code)) |
||||
return r.text |
||||
except KeyboardInterrupt: |
||||
raise |
||||
except BaseException as e: |
||||
logging.error('Problem hitting URL: {}'.format(str(e))) |
||||
return False |
||||
|
||||
def get_article_details(url): |
||||
try: |
||||
params = {'source_url': url} |
||||
headers = {'Referer': 'https://outline.com/'} |
||||
r = requests.get(OUTLINE_API, params=params, headers=headers, timeout=20) |
||||
if r.status_code == 429: |
||||
logging.info('Rate limited by outline, sleeping 30s and skipping...') |
||||
time.sleep(30) |
||||
return '' |
||||
if r.status_code != 200: |
||||
raise Exception('Bad response code ' + str(r.status_code)) |
||||
data = r.json()['data'] |
||||
if 'URL is not supported by Outline' in data['html']: |
||||
raise Exception('URL not supported by Outline') |
||||
return (data, "outline") |
||||
except KeyboardInterrupt: |
||||
raise |
||||
except BaseException as e: |
||||
logging.error('Problem outlining article: {}'.format(str(e))) |
||||
return (None, None) |
||||
|
||||
|
||||
class Sitemap: |
||||
def __init__(self, url): |
||||
self.sitemap_url = url |
||||
|
||||
def feed(self): |
||||
markup = xml(lambda x: self.sitemap_url) |
||||
if not markup: return [] |
||||
soup = BeautifulSoup(markup, features='lxml') |
||||
articles = soup.find('urlset').findAll('url') |
||||
articles = list(filter(None, [a if a.find('lastmod') is not None else None for a in articles])) |
||||
return [x.find('loc').text for x in articles] or [] |
||||
|
||||
def story(self, ref): |
||||
markup = xml(lambda x: self.sitemap_url) |
||||
if not markup: return [] |
||||
soup = BeautifulSoup(markup, features='lxml') |
||||
articles = soup.find('urlset').findAll('url') |
||||
articles = list(filter(None, [a if a.find('lastmod') is not None else None for a in articles])) |
||||
articles = list(filter(None, [a if a.find('loc').text == ref else None for a in articles])) |
||||
|
||||
if len(articles) == 0: |
||||
return False |
||||
|
||||
r = articles[0] |
||||
if not r: |
||||
return False |
||||
|
||||
(data, method) = get_article_details(ref) |
||||
if not data: |
||||
return False |
||||
if 'outline' not in method: |
||||
return False |
||||
s = {} |
||||
s['author'] = data['author'] |
||||
s['author_link'] = '' |
||||
s['date'] = unix(r.find('lastmod').text) |
||||
s['score'] = 0 |
||||
s['title'] = data['title'] |
||||
s['link'] = data['article_url'] |
||||
s['url'] = data['article_url'] |
||||
s['comments'] = [] |
||||
s['num_comments'] = 0 |
||||
s['text'] = data['html'] |
||||
|
||||
return s |
||||
|
||||
|
||||
# scratchpad so I can quickly develop the parser |
||||
if __name__ == '__main__': |
||||
# site = Sitemap("https://www.stuff.co.nz/sitemap.xml") |
||||
site = Sitemap("https://www.nzherald.co.nz/arcio/news-sitemap/") |
||||
posts = site.feed() |
||||
print(posts[:1]) |
||||
print(site.story(posts[0])) |
Loading…
Reference in new issue