mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-06-23 15:30:45 -04:00
207 lines
8.5 KiB
Python
207 lines
8.5 KiB
Python
#!/usr/bin/env python
|
|
# vim:fileencoding=utf-8
|
|
import json
|
|
import time
|
|
from datetime import datetime, timedelta
|
|
from itertools import zip_longest
|
|
|
|
from calibre.ebooks.BeautifulSoup import BeautifulSoup
|
|
from calibre.web.feeds.news import BasicNewsRecipe, classes
|
|
|
|
|
|
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
|
|
|
|
recipe_specific_options = {
|
|
'date': {
|
|
'short': 'The date of the edition to download (YYYYMMDD format)\nOnly the past 6 editions will be available ',
|
|
'long': 'For example, 20240513'
|
|
},
|
|
'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):
|
|
d = self.recipe_specific_options.get('date')
|
|
if not (d and isinstance(d, str)):
|
|
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')), # noqa: ISC001
|
|
]
|
|
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 = [''.join([n for n in itm['key'] if n.isdigit()]) for itm in catalog['items'] if itm['type'] == 'ITP'][1:]
|
|
self.log('**Past Editions available :', ', '.join(edit))
|
|
|
|
past_edition = self.recipe_specific_options.get('date')
|
|
|
|
for itm in catalog['items']:
|
|
if past_edition and isinstance(past_edition, str):
|
|
if past_edition in itm['key']:
|
|
manifest = itm['manifest']
|
|
date = itm['date']
|
|
break
|
|
elif itm['type'] == 'ITP':
|
|
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('_', ' ')
|
|
if 'MAGAZINE' in section:
|
|
if not (dt.day in {1, 2, 3, 4, 5, 6, 7} and dt.weekday() == 5):
|
|
continue
|
|
self.log('Loading Magazine section')
|
|
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']
|
|
url = index + manifest.rsplit('/', 1)[0] + '/' + 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):
|
|
return BeautifulSoup(raw).prettify()
|
|
|
|
def populate_article_metadata(self, article, soup, first):
|
|
lnk = soup.find('div', attrs={'id':'share-link'})
|
|
if lnk:
|
|
article.url = lnk['title']
|