| 1 |
#!/usr/bin/env python |
| 2 |
# vim:fileencoding=utf-8 |
| 3 |
""" |
| 4 |
www.washingtonpost.com |
| 5 |
""" |
| 6 |
|
| 7 |
import json |
| 8 |
|
| 9 |
from html5_parser import parse |
| 10 |
|
| 11 |
from calibre.web.feeds.news import BasicNewsRecipe |
| 12 |
|
| 13 |
|
| 14 |
class TheWashingtonPost(BasicNewsRecipe): |
| 15 |
title = 'The Washington Post' |
| 16 |
__author__ = 'unkn0wn' |
| 17 |
description = ( |
| 18 |
'Leading source for news, video and opinion on politics, business, ' |
| 19 |
'world and national news, science, travel, entertainment and more. ' |
| 20 |
'Our local coverage includes reporting on education, crime, weather, ' |
| 21 |
'traffic, real estate, jobs and cars for DC, Maryland and Virginia. ' |
| 22 |
'Offering award-winning opinion writing, entertainment information ' |
| 23 |
'and restaurant reviews.' |
| 24 |
) |
| 25 |
publisher = 'The Washington Post Company' |
| 26 |
category = 'news, politics, USA' |
| 27 |
oldest_article = 1.2 |
| 28 |
max_articles_per_feed = 200 |
| 29 |
no_stylesheets = True |
| 30 |
encoding = 'utf8' |
| 31 |
use_embedded_content = False |
| 32 |
language = 'en_US' |
| 33 |
remove_empty_feeds = True |
| 34 |
resolve_internal_links = True |
| 35 |
ignore_duplicate_articles = {'url'} |
| 36 |
masthead_url = 'https://upload.wikimedia.org/wikipedia/commons/9/93/The_Logo_of_The_Washington_Post_Newspaper.svg' |
| 37 |
publication_type = 'newspaper' |
| 38 |
remove_attributes = ['style', 'width', 'height'] |
| 39 |
|
| 40 |
recipe_specific_options = { |
| 41 |
'days': { |
| 42 |
'short': 'Oldest article to download from this news source. In days ', |
| 43 |
'long': 'For example, 0.5, gives you articles from the past 12 hours', |
| 44 |
'default': str(oldest_article), |
| 45 |
} |
| 46 |
} |
| 47 |
|
| 48 |
def __init__(self, *args, **kwargs): |
| 49 |
BasicNewsRecipe.__init__(self, *args, **kwargs) |
| 50 |
d = self.recipe_specific_options.get('days') |
| 51 |
if d and isinstance(d, str): |
| 52 |
self.oldest_article = float(d) |
| 53 |
|
| 54 |
extra_css = ''' |
| 55 |
.img { text-align:center; font-size:small; } |
| 56 |
.auth { font-weight:bold; font-size:small; } |
| 57 |
.time { font-size:small; color: #202020; } |
| 58 |
.subt { font-style: italic; } |
| 59 |
''' |
| 60 |
|
| 61 |
def get_cover_url(self): |
| 62 |
soup = self.index_to_soup('https://www.washingtonpost.com/todays_paper/updates/') |
| 63 |
if img := soup.find('img', attrs={'src': lambda x: x and x.endswith('_FrontPage.png')}): |
| 64 |
return img['src'] |
| 65 |
|
| 66 |
def get_browser(self, *args, **kwargs): |
| 67 |
kwargs['user_agent'] = 'Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)' |
| 68 |
br = BasicNewsRecipe.get_browser(self, *args, **kwargs) |
| 69 |
br.addheaders += [ |
| 70 |
('Referer', 'https://www.google.com/'), |
| 71 |
('X-Forwarded-For', '66.249.66.1'), |
| 72 |
] |
| 73 |
return br |
| 74 |
|
| 75 |
# Official feeds: https://www.washingtonpost.com/discussions/2018/10/12/washington-post-rss-feeds/ |
| 76 |
feeds = [ |
| 77 |
('Politics', 'http://feeds.washingtonpost.com/rss/politics'), |
| 78 |
('Opinions', 'http://feeds.washingtonpost.com/rss/opinions'), |
| 79 |
('Local', 'http://feeds.washingtonpost.com/rss/local'), |
| 80 |
('Sports', 'http://feeds.washingtonpost.com/rss/sports'), |
| 81 |
('Technology', 'http://feeds.washingtonpost.com/rss/business/technology'), |
| 82 |
('National', 'http://feeds.washingtonpost.com/rss/national'), |
| 83 |
('World', 'http://feeds.washingtonpost.com/rss/world'), |
| 84 |
('Business', 'http://feeds.washingtonpost.com/rss/business'), |
| 85 |
('Lifestyle', 'http://feeds.washingtonpost.com/rss/lifestyle'), |
| 86 |
('Entertainment', 'http://feeds.washingtonpost.com/rss/entertainment'), |
| 87 |
# Undocumented feeds. |
| 88 |
('White House', 'http://feeds.washingtonpost.com/rss/politics/whitehouse'), |
| 89 |
('Commanders', 'http://feeds.washingtonpost.com/rss/sports/redskins'), |
| 90 |
] |
| 91 |
|
| 92 |
def preprocess_raw_html(self, raw, url): |
| 93 |
root = parse(raw) |
| 94 |
if '/interactive/' in url: |
| 95 |
return ( |
| 96 |
'<html><body><article><h1>' + root.xpath('//h1')[0].text + '</h1><em>' |
| 97 |
'This article is supposed to be read in a browser.' |
| 98 |
'</em></article></body></html>' |
| 99 |
) |
| 100 |
m = root.xpath('//script[@id="__NEXT_DATA__"]') |
| 101 |
|
| 102 |
data = json.loads(m[0].text) |
| 103 |
data = data['props']['pageProps']['globalContent'] |
| 104 |
|
| 105 |
text = data.get('label', {}).get('basic', {}).get('text', '') |
| 106 |
label = f'<p class="time">{text}</p>' if text else '' |
| 107 |
title = '<h1>' + data['headlines']['basic'] + '</h1>' |
| 108 |
subhead = '<p class="subt">' + data['description'].get('basic', '') + '</h3>' |
| 109 |
|
| 110 |
promo_img = '' |
| 111 |
if data.get('promo_items', {}).get('basic', {}).get('type', '') == 'image': |
| 112 |
pi = data['promo_items']['basic'] |
| 113 |
promo_img = '<p><div class="img"><img src="{}"><div>{}</div></div></p>'.format(pi['url'], pi['credits_caption_display']) |
| 114 |
|
| 115 |
author = '' |
| 116 |
if 'credits' in data: |
| 117 |
author = ( |
| 118 |
'<div><span class="auth">' |
| 119 |
+ 'By ' |
| 120 |
+ ', '.join(x['name'] for x in data['credits']['by']) |
| 121 |
+ '</span> | <span class="time">' |
| 122 |
+ data['publish_date'][:-14] |
| 123 |
+ '</span></div>' |
| 124 |
) |
| 125 |
|
| 126 |
body = '' |
| 127 |
for x in data['content_elements']: |
| 128 |
if x['type'] == 'text': |
| 129 |
body += '<p>' + x['content'] + '</p>' |
| 130 |
elif x['type'] == 'video': |
| 131 |
if 'promo_image' in x: |
| 132 |
body += '<p><div class="img"><img src="{}"><div>{}</div></div></p>'.format(x['promo_image']['url'], x['description'].get('basic', '')) |
| 133 |
elif x['type'] == 'image': |
| 134 |
img_ = '<p><div class="img"><img src="{}"><div>{}</div></div></p>'.format(x['url'], x['credits_caption_display']) |
| 135 |
if img_ != promo_img: |
| 136 |
body += img_ |
| 137 |
elif x['type'] == 'list': |
| 138 |
body += '<ul>' |
| 139 |
for li in x['items']: |
| 140 |
if li.get('content', '') != '': |
| 141 |
body += f'<li>{li["content"]}</li>' |
| 142 |
body += '</ul>' |
| 143 |
|
| 144 |
return '<html><body><div>' + label + title + subhead + promo_img + author + body + '</div></body></html>' |
| 145 |
|
| 146 |
def preprocess_html(self, soup): |
| 147 |
for img in soup.findAll('img', attrs={'src': True}): |
| 148 |
img['src'] = 'https://www.washingtonpost.com/wp-apps/imrs.php?src=' + img['src'] + '&w=600' |
| 149 |
return soup |