calibre/recipes/wsj.recipe
unkn_wn 72cba9ee74
Update wsj.recipe
change cover source
2025-07-29 14:40:17 +05:30

276 lines
10 KiB
Python

#!/usr/bin/env python
# vim:fileencoding=utf-8
import json
import time
from itertools import zip_longest
from urllib.parse import quote, urlencode
from calibre import browser
from calibre.ebooks.BeautifulSoup import BeautifulSoup
from calibre.ptempfile import PersistentTemporaryFile
from calibre.web.feeds.news import BasicNewsRecipe, classes
def get_article(article_id):
from mechanize import Request
mat_url = 'https://mats.mobile.dowjones.io/translate/' + article_id + '/jpml'
headers = {
'User-Agent': 'okhttp/4.10.0',
'Accept-Encoding': 'gzip',
'Cache-Control': 'no-cache',
'x-api-key': ('e''0''5''9''9''5''f''f''4''4''2''1''4''3''2''5''5''e''b''8''3''8''1''f''7''2''d''4''9''1''3''b''f''7''5''0''3''d''6''c'), # noqa: ISC001
}
br = browser()
req = Request(
mat_url,
headers=headers,
)
res = br.open(req)
return res.read()
class WSJ(BasicNewsRecipe):
title = 'The Wall Street Journal'
__author__ = 'unkn0wn'
description = (
'The Print Edition of WSJ. 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
simultaneous_downloads = 20
recipe_specific_options = {
'date': {
'short': 'The date of the edition to download (YYYY-MM-DD format)\nOnly the past 6 editions will be available ',
'long': 'For example, 2024-05-13',
},
'res': {
'short': 'For hi-res images, select a resolution from the\nfollowing options: 800, 1000, 1200 or 1500',
'long': 'This is useful for non e-ink devices, and for a lower file size\nthan the default, use 400 or 300.',
'default': '600',
},
}
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='panel', attrs={'id': 'summary-image'}),
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',
'article-manifest',
'body-extract',
'category',
'sub-category',
'socialhed',
'summary',
'deckline',
'article-flashline',
]
},
),
]
remove_tags_before = [dict(name='p', attrs={'id': 'orig-pubdate-string'})]
def media_bucket(self, x):
res = '?width=600'
w = self.recipe_specific_options.get('res')
if w and isinstance(w, str):
res = '?width=' + w
if x.get('type', '') == 'image':
if (
x.get('subtype', '') == 'graphic'
or 'images.wsj.net' not in x['manifest-url']
):
return '<br><img src="{}"><div class="figc">{}</div>\n'.format(
x['manifest-url'], x['caption'] + '<i> ' + x['credit'] + '</i>'
)
return '<br><img src="{}"><div class="figc">{}</div>\n'.format(
x['manifest-url'].split('?')[0] + res,
x['caption'] + '<i> ' + x['credit'] + '</i>',
)
if x.get('type', '') == 'video':
return (
'<br><a href="{}"><img src="{}"></a><div class="figc">{}</div>\n'.format(
x['share_link'],
x['thumbnail_url'].split('?')[0] + res,
x['caption'] + '<i> ' + x['credit'] + '</i>',
)
)
return
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'})
fl = soup.find('p', attrs={'id': 'flashline'})
if dt and byl and read and fl:
dt.name = read.name = byl.name = fl.name = 'div'
byl.insert(0, dt)
byl.insert(0, read)
url = soup.find('p', attrs={'id': 'share-link'})
if url:
url.name = 'div'
url['title'] = self.tag_to_string(url).strip()
url.string = ''
panel = soup.find('panel', attrs={'id': 'metadata'})
if panel:
buck = panel.find('p', attrs={'id': 'media-bucket'})
if buck:
data = json.loads(buck.string)
buck.extract()
i_lst = [self.media_bucket(x) for x in data['items']]
m_itm = soup.findAll('panel', attrs={'class': 'media-item'})
if i_lst and m_itm:
for x, y in list(zip_longest(m_itm, i_lst)):
x.insert_after(BeautifulSoup(y, 'html.parser'))
return soup
def postprocess_html(self, soup, first_fetch):
for pan in soup.findAll('panel'):
pan.name = 'div'
return soup
def _download_cover(self):
import os
from contextlib import closing
from calibre.utils.img import save_cover_data_to
br = browser()
dt = self.recipe_specific_options.get('date')
if (dt and isinstance(dt, str)):
d, m, y = dt.split('-')
cu = f'https://www.wsj.com/public/resources/documents/WSJNewsPaper-{int(m)}-{int(d)}-{y}.jpg'
else:
raw = br.open('https://frontpages.freedomforum.org/newspapers/wsj-The_Wall_Street_Journal')
soup = BeautifulSoup(raw.read())
cu = soup.find(
'img',
attrs={
'alt': 'Front Page Image',
'src': lambda x: x and x.endswith('front-page-large.jpg'),
},
)['src'].replace('-large', '-medium')
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):
br = BasicNewsRecipe.get_browser(self, *args, **kw)
br.addheaders += [
('apollographql-client-name', 'wsj-mobile-android-release'),
]
return br
def parse_index(self):
query = {
'operationName': 'IssueQuery',
'variables': '{"publication":"WSJ","region":"US","masthead":"ITPNEXTGEN"}',
'extensions': '{"persistedQuery":{"version":1,"sha256Hash":"d938226e7d1c1fff050e7d084c72179e2713dcf4736d3a442c618c55b896f847"}}',
}
url = 'https://shared-data.dowjones.io/gateway/graphql?' + urlencode(
query, safe='()!', quote_via=quote
)
raw = self.index_to_soup(url, raw=True)
cat_data = json.loads(raw)['data']['mobileIssuesByMasthead']
edit = [x['datedLabel'] for x in cat_data][1:]
self.log('**Past Editions available : ' + ' | '.join(edit))
past_edition = self.recipe_specific_options.get('date')
for itm in cat_data:
if past_edition and isinstance(past_edition, str):
if past_edition in itm['publishedDateUtc']:
self.timefmt = ' [' + itm['datedLabel']
sections_ = itm['sections']
break
self.timefmt = f' [{itm["datedLabel"]}]'
sections_ = itm['sections']
break
self.log('Downloading ', self.timefmt)
feeds = []
for sec in sections_[:-1]:
time.sleep(3)
section = sec['label']
self.log(section)
cont_id = sec['key']
query = {
'operationName': 'SectionQuery',
'variables': '{{"id":"{}"}}'.format(cont_id),
'extensions': '{"persistedQuery":{"version":1,"sha256Hash":"207fe93376f379bf223ed2734cf9313a28291293366a803db923666fa6b45026"}}',
}
sec_url = 'https://shared-data.dowjones.io/gateway/graphql?' + urlencode(
query, safe='()!', quote_via=quote
)
sec_raw = self.index_to_soup(sec_url, raw=True)
sec_data = json.loads(sec_raw)['data']['summaryCollectionContent'][
'collectionItems'
]
articles = []
for art in sec_data:
for arts in art['collectionItems']:
mobi = arts['content']['mobileSummary']
title = mobi['headline']['text']
desc = mobi['description']['content']['text']
url = arts['id']
self.log(' ', title, '\n\t', desc)
articles.append({'title': title, 'description': desc, 'url': url})
feeds.append((section, articles))
return feeds
def populate_article_metadata(self, article, soup, first):
lnk = soup.find('div', attrs={'id': 'share-link'})
if lnk:
article.url = lnk['title']
def print_version(self, url):
art_cont = get_article(url)
pt = PersistentTemporaryFile('.html')
pt.write(art_cont)
pt.close()
return 'file:///' + pt.name