163 lines
4.8 KiB
Python
163 lines
4.8 KiB
Python
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 utils import clean
|
|
|
|
API_TOPSTORIES = lambda x: 'https://hacker-news.firebaseio.com/v0/topstories.json'
|
|
ALG_API_ITEM = lambda x : 'https://hn.algolia.com/api/v1/items/{}'.format(x)
|
|
BHN_API_ITEM = lambda x : 'https://api.hnpwa.com/v0/item/{}.json'.format(x)
|
|
|
|
SITE_LINK = lambda x : 'https://news.ycombinator.com/item?id={}'.format(x)
|
|
SITE_AUTHOR_LINK = lambda x : 'https://news.ycombinator.com/user?id={}'.format(x)
|
|
|
|
def api(route, ref=None):
|
|
try:
|
|
r = requests.get(route(ref), timeout=5)
|
|
if r.status_code != 200:
|
|
raise Exception('Bad response code ' + str(r.status_code))
|
|
return r.json()
|
|
except KeyboardInterrupt:
|
|
raise
|
|
except BaseException as e:
|
|
logging.error('Problem hitting hackernews API: {}, trying again'.format(str(e)))
|
|
|
|
try:
|
|
r = requests.get(route(ref), timeout=15)
|
|
if r.status_code != 200:
|
|
raise Exception('Bad response code ' + str(r.status_code))
|
|
return r.json()
|
|
except KeyboardInterrupt:
|
|
raise
|
|
except BaseException as e:
|
|
logging.error('Problem hitting hackernews API: {}'.format(str(e)))
|
|
return False
|
|
|
|
def feed():
|
|
return [str(x) for x in api(API_TOPSTORIES) or []]
|
|
|
|
def alg_comment(i):
|
|
if 'author' not in i:
|
|
return False
|
|
|
|
c = {}
|
|
c['author'] = i.get('author', '')
|
|
c['score'] = i.get('points', 0)
|
|
c['date'] = i.get('created_at_i', 0)
|
|
c['text'] = clean(i.get('text', '') or '')
|
|
c['comments'] = [alg_comment(j) for j in i['children']]
|
|
c['comments'] = list(filter(bool, c['comments']))
|
|
return c
|
|
|
|
def alg_comment_count(i):
|
|
alive = 1 if i['author'] else 0
|
|
return sum([alg_comment_count(c) for c in i['comments']]) + alive
|
|
|
|
def alg_story(ref):
|
|
r = api(ALG_API_ITEM, ref)
|
|
if not r:
|
|
logging.info('Bad Algolia Hackernews API response.')
|
|
return None
|
|
|
|
if 'deleted' in r:
|
|
logging.info('Story was deleted.')
|
|
return False
|
|
elif r.get('type', '') != 'story':
|
|
logging.info('Type "{}" is not "story".'.format(r.get('type', '')))
|
|
return False
|
|
|
|
s = {}
|
|
s['author'] = r.get('author', '')
|
|
s['author_link'] = SITE_AUTHOR_LINK(r.get('author', ''))
|
|
s['score'] = r.get('points', 0)
|
|
s['date'] = r.get('created_at_i', 0)
|
|
s['title'] = r.get('title', '')
|
|
s['link'] = SITE_LINK(ref)
|
|
s['url'] = r.get('url', '')
|
|
s['comments'] = [alg_comment(i) for i in r['children']]
|
|
s['comments'] = list(filter(bool, s['comments']))
|
|
s['num_comments'] = alg_comment_count(s) - 1
|
|
|
|
if 'text' in r and r['text']:
|
|
s['text'] = clean(r['text'] or '')
|
|
|
|
return s
|
|
|
|
def bhn_comment(i):
|
|
if 'user' not in i:
|
|
return False
|
|
|
|
c = {}
|
|
c['author'] = i.get('user', '')
|
|
c['score'] = 0 # Not present?
|
|
c['date'] = i.get('time', 0)
|
|
c['text'] = clean(i.get('content', '') or '')
|
|
c['comments'] = [bhn_comment(j) for j in i['comments']]
|
|
c['comments'] = list(filter(bool, c['comments']))
|
|
return c
|
|
|
|
def bhn_story(ref):
|
|
r = api(BHN_API_ITEM, ref)
|
|
if not r:
|
|
logging.info('Bad BetterHN Hackernews API response.')
|
|
return None
|
|
|
|
if 'deleted' in r: # TODO: verify
|
|
logging.info('Story was deleted.')
|
|
return False
|
|
elif r.get('type', '') != 'link':
|
|
logging.info('Type "{}" is not "link".'.format(r.get('type', '')))
|
|
return False
|
|
|
|
s = {}
|
|
s['author'] = r.get('user', '')
|
|
s['author_link'] = SITE_AUTHOR_LINK(r.get('user', ''))
|
|
s['score'] = r.get('points', 0)
|
|
s['date'] = r.get('time', 0)
|
|
s['title'] = r.get('title', '')
|
|
s['link'] = SITE_LINK(ref)
|
|
s['url'] = r.get('url', '')
|
|
if s['url'].startswith('item'):
|
|
s['url'] = SITE_LINK(ref)
|
|
s['comments'] = [bhn_comment(i) for i in r['comments']]
|
|
s['comments'] = list(filter(bool, s['comments']))
|
|
s['num_comments'] = r.get('comments_count', 0)
|
|
|
|
if 'content' in r and r['content']:
|
|
s['text'] = clean(r['content'] or '')
|
|
|
|
return s
|
|
|
|
def story(ref):
|
|
s = alg_story(ref)
|
|
if s is None:
|
|
s = bhn_story(ref)
|
|
if not s:
|
|
return False
|
|
|
|
|
|
if s['score'] < 25 and s['num_comments'] < 10:
|
|
logging.info('Score ({}) or num comments ({}) below threshold.'.format(s['score'], s['num_comments']))
|
|
return False
|
|
|
|
return s
|
|
|
|
# scratchpad so I can quickly develop the parser
|
|
if __name__ == '__main__':
|
|
print(feed())
|
|
#print(story(20763961))
|
|
#print(story(20802050))
|
|
|
|
#print(story(42899834)) # type "job"
|
|
#print(story(42900076)) # Ask HN
|
|
#print(story(42898201)) # Show HN
|
|
#print(story(42899703)) # normal
|
|
print(story(42902678)) # bad title?
|