mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-10-30 18:22:25 -04:00
200 lines
7.6 KiB
Python
200 lines
7.6 KiB
Python
import json
|
|
import re
|
|
import time
|
|
from datetime import datetime, timedelta
|
|
|
|
from calibre.ebooks.BeautifulSoup import BeautifulSoup
|
|
from calibre.web.feeds.news import BasicNewsRecipe, classes
|
|
from html5_parser import parse
|
|
|
|
# Past 6 editions are available for download.
|
|
# For available past editions see log and set date to, for example, '20240513'.
|
|
past_edition = None
|
|
|
|
|
|
class WSJ(BasicNewsRecipe):
|
|
title = 'The Wall Street Journal'
|
|
__author__ = 'unkn0wn'
|
|
description = (
|
|
'The Wall Street Journal is your source for breaking news, analysis and insights from the U.S. and '
|
|
'around the world, the world\'s leading business and finance publication.'
|
|
)
|
|
language = 'en_US'
|
|
masthead_url = 'https://s.wsj.net/media/wsj_amp_masthead_lg.png'
|
|
encoding = 'utf-8'
|
|
no_javascript = True
|
|
no_stylesheets = True
|
|
remove_attributes = ['style', 'height', 'width']
|
|
resolve_internal_links = True
|
|
|
|
extra_css = '''
|
|
#subhed, em { font-style:italic; color:#202020; }
|
|
#byline, #time-to-read, #orig-pubdate-string, .article-byline, time, #flashline { font-size:small; }
|
|
.figc { font-size:small; text-align:center; }
|
|
img {display:block; margin:0 auto;}
|
|
'''
|
|
|
|
remove_tags = [
|
|
dict(name=['nav', 'svg', 'iframe', 'source']),
|
|
dict(name='panel', attrs={'id':'metadata'}),
|
|
dict(name='panel', attrs={'layout':'inline'}),
|
|
dict(name='panel', attrs={'embed':'inner-article-ad'}),
|
|
dict(name='span', attrs={'embed':'ticker'}),
|
|
classes('lamrelated-articles-inset-panel'),
|
|
dict(name='p', attrs={'id':[
|
|
'keywords', 'orig-pubdate-number', 'type', 'is-custom-flashline', 'grouphed', 'author-ids',
|
|
'body-extract', 'category', 'sub-category', 'socialhed', 'summary', 'deckline', 'article-flashline'
|
|
]}),
|
|
dict(attrs={'data-inset_type':'dynamic'}),
|
|
dict(attrs={'data-block':'dynamic-inset'})
|
|
]
|
|
|
|
remove_tags_before = [
|
|
dict(name='p', attrs={'id':'orig-pubdate-string'})
|
|
]
|
|
remove_tags_after = [
|
|
dict(name='article')
|
|
]
|
|
|
|
def preprocess_html(self, soup):
|
|
jpml = soup.find('jpml')
|
|
if jpml:
|
|
jpml.name = 'article'
|
|
h1 = soup.find('p', attrs={'id':'headline'})
|
|
if h1:
|
|
h1.name = 'h1'
|
|
for h2 in soup.findAll('h2'):
|
|
h2.name = 'h4'
|
|
dt = soup.find('p', attrs={'id':'orig-pubdate-string'})
|
|
read = soup.find('p', attrs={'id':'time-to-read'})
|
|
byl = soup.find('p', attrs={'id':'byline'})
|
|
if dt and byl and read:
|
|
dt.name = read.name = byl.name = 'div'
|
|
byl.insert(0, dt)
|
|
byl.insert(0, read)
|
|
url = soup.find('p', attrs={'id':'share-link'})
|
|
if url:
|
|
url['title'] = self.tag_to_string(url).strip()
|
|
url.string = ''
|
|
for img in soup.findAll('img', attrs={'location':True}):
|
|
img['src'] = img['location']
|
|
for figc in soup.findAll('figcaption'):
|
|
figc['class'] = 'figc'
|
|
col = soup.find('div', text = re.compile('What to Read Next'))
|
|
if col:
|
|
div = col.findParent('div')
|
|
if div:
|
|
div.extract()
|
|
time = soup.find('time')
|
|
if time:
|
|
p = time.findParent('div')
|
|
if p:
|
|
p.name = 'p'
|
|
for img in soup.findAll('img', src=True):
|
|
if img['src'].endswith('/OR'):
|
|
img['src'] = img['src'][:-3]
|
|
return soup
|
|
|
|
if not past_edition:
|
|
def _download_cover(self):
|
|
import os
|
|
from contextlib import closing
|
|
|
|
from calibre import browser
|
|
from calibre.utils.img import save_cover_data_to
|
|
br = browser()
|
|
raw = br.open('https://www.frontpages.com/the-wall-street-journal/')
|
|
soup = BeautifulSoup(raw.read())
|
|
cu = 'https://www.frontpages.com' + soup.find('img', attrs={'id':'giornale-img'})['src']
|
|
self.report_progress(1, _('Downloading cover from %s')%cu)
|
|
with closing(br.open(cu, timeout=self.timeout)) as r:
|
|
cdata = r.read()
|
|
cpath = os.path.join(self.output_dir, 'cover.jpg')
|
|
save_cover_data_to(cdata, cpath)
|
|
self.cover_path = cpath
|
|
|
|
def get_browser(self, *args, **kw):
|
|
kw['user_agent'] = 'okhttp/4.10.0'
|
|
br = BasicNewsRecipe.get_browser(self, *args, **kw)
|
|
br.addheaders += [
|
|
('Accept-Encoding', 'gzip'),
|
|
('cache-control', 'no-cache'),
|
|
('x-api-key', ('e''b''2''4''0''8''c''d''2''7''f''8''9''1''3''d''4''2''1''f''a''3''d''5''c''3''d''0''7''c''c''f''0''3''4''c''b''4''4''8')),
|
|
]
|
|
return br
|
|
|
|
def parse_index(self):
|
|
index = 'https://bartender.mobile.dowjones.io'
|
|
catalog = json.loads(self.index_to_soup(index + '/catalogs/v1/wsj/us/catalog.json', raw=True))
|
|
edit = [itm['key'][3:] for itm in catalog['items'] if itm['type'] == 'ITP'][1:]
|
|
self.log('**Past Editions available :', ', '.join(edit))
|
|
for itm in catalog['items']:
|
|
if past_edition:
|
|
if itm['key'] == 'ITP' + past_edition:
|
|
key = itm['key']
|
|
manifest = itm['manifest']
|
|
date = itm['date']
|
|
break
|
|
elif itm['type'] == 'ITP':
|
|
key = itm['key']
|
|
manifest = itm['manifest']
|
|
date = itm['date']
|
|
break
|
|
|
|
dt = datetime.fromisoformat(date[:-1]) + timedelta(seconds=time.timezone)
|
|
dt = dt.strftime('%b %d, %Y')
|
|
self.log('Downloading ', dt)
|
|
self.timefmt = ' [' + dt + ']'
|
|
|
|
feeds = []
|
|
|
|
manif = json.loads(self.index_to_soup(index + manifest, raw=True))
|
|
for itm in manif['items']:
|
|
for k, v in itm.items():
|
|
if '-pages_' in k:
|
|
section = k.split('-pages_')[0].replace('_', ' ')
|
|
self.log(section)
|
|
|
|
articles = []
|
|
|
|
sec_parse = json.loads(self.index_to_soup(index + v, raw=True))
|
|
data = sec_parse['articles']
|
|
for art in data:
|
|
title = data[art]['headline']
|
|
desc = data[art]['summary']
|
|
if 'articleWebViewLink' in data[art]:
|
|
url = data[art]['articleWebViewLink']
|
|
else:
|
|
url = index + '/contents/v1/wsj/us/' + key + '/' + data[art]['filename']
|
|
self.log(' ', title, '\n\t', desc)
|
|
articles.append({'title': title, 'description':desc, 'url': url})
|
|
feeds.append((section, articles))
|
|
return feeds
|
|
|
|
def preprocess_raw_html(self, raw, url):
|
|
if '/webview/' not in url:
|
|
root = parse(raw)
|
|
for x in root.xpath('//image'):
|
|
x.tag = 'img'
|
|
return BeautifulSoup(raw).prettify()
|
|
else:
|
|
soup = BeautifulSoup(raw)
|
|
url = soup.find('meta', attrs={'property':'og:url'})
|
|
if url:
|
|
h1 = soup.find('h1')
|
|
if h1:
|
|
h1['title'] = url['content']
|
|
h2 = soup.find('h2')
|
|
if h2:
|
|
h2['id'] = 'subhed'
|
|
h2.name = 'p'
|
|
return soup.prettify()
|
|
|
|
def populate_article_metadata(self, article, soup, first):
|
|
lnk = soup.find('p', attrs={'id':'share-link'})
|
|
if lnk:
|
|
article.url = lnk['title']
|
|
art = soup.find('h1', attrs={'title':True})
|
|
if art:
|
|
article.url = art['title']
|