mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
Update wsj.recipe
This commit is contained in:
unkn0w7n
parent
49934879c7
commit
a0a1b8a8f3
@ -1,10 +1,12 @@
|
||||
#!/usr/bin/env python
|
||||
# vim:fileencoding=utf-8
|
||||
import json
|
||||
import time
|
||||
from datetime import datetime, timedelta
|
||||
from itertools import zip_longest
|
||||
from urllib.parse import quote, urlencode
|
||||
from mechanize import Request
|
||||
|
||||
from calibre.ptempfile import PersistentTemporaryFile
|
||||
from calibre.ebooks.BeautifulSoup import BeautifulSoup
|
||||
from calibre.web.feeds.news import BasicNewsRecipe, classes
|
||||
|
||||
@ -23,17 +25,18 @@ class WSJ(BasicNewsRecipe):
|
||||
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 (YYYYMMDD format)\nOnly the past 6 editions will be available ',
|
||||
'long': 'For example, 20240513'
|
||||
'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'
|
||||
}
|
||||
'default': '600',
|
||||
},
|
||||
}
|
||||
|
||||
extra_css = '''
|
||||
@ -44,20 +47,35 @@ class WSJ(BasicNewsRecipe):
|
||||
'''
|
||||
|
||||
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'}),
|
||||
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'
|
||||
]}),
|
||||
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'})
|
||||
]
|
||||
remove_tags_before = [dict(name='p', attrs={'id': 'orig-pubdate-string'})]
|
||||
|
||||
def media_bucket(self, x):
|
||||
res = '?width=600'
|
||||
@ -65,16 +83,24 @@ class WSJ(BasicNewsRecipe):
|
||||
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']:
|
||||
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>'
|
||||
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 (
|
||||
'<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
|
||||
|
||||
@ -82,32 +108,32 @@ class WSJ(BasicNewsRecipe):
|
||||
jpml = soup.find('jpml')
|
||||
if jpml:
|
||||
jpml.name = 'article'
|
||||
h1 = soup.find('p', attrs={'id':'headline'})
|
||||
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'})
|
||||
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'})
|
||||
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'})
|
||||
panel = soup.find('panel', attrs={'id': 'metadata'})
|
||||
if panel:
|
||||
buck = panel.find('p', attrs={'id':'media-bucket'})
|
||||
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'})
|
||||
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'))
|
||||
@ -126,11 +152,15 @@ class WSJ(BasicNewsRecipe):
|
||||
|
||||
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)
|
||||
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')
|
||||
@ -138,69 +168,102 @@ class WSJ(BasicNewsRecipe):
|
||||
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
|
||||
('apollographql-client-name', 'wsj-mobile-android-release'),
|
||||
]
|
||||
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))
|
||||
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)
|
||||
|
||||
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 catalog['items']:
|
||||
for itm in cat_data:
|
||||
if past_edition and isinstance(past_edition, str):
|
||||
if past_edition in itm['key']:
|
||||
manifest = itm['manifest']
|
||||
date = itm['date']
|
||||
if past_edition in itm['publishedDateUtc']:
|
||||
self.timefmt = ' [' + itm['datedLabel']
|
||||
sections_ = itm['sections']
|
||||
break
|
||||
elif itm['type'] == 'ITP':
|
||||
manifest = itm['manifest']
|
||||
date = itm['date']
|
||||
break
|
||||
self.timefmt = f' [{itm["datedLabel"]}]'
|
||||
sections_ = itm['sections']
|
||||
break
|
||||
|
||||
dt = datetime.fromisoformat(date[:-1]) + timedelta(seconds=time.timezone)
|
||||
dt_ = dt.strftime('%b %d, %Y')
|
||||
self.log('Downloading ', dt_)
|
||||
self.timefmt = ' [' + dt_ + ']'
|
||||
self.log('Downloading ', self.timefmt)
|
||||
|
||||
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)
|
||||
for sec in sections_[:-1]:
|
||||
section = sec['label']
|
||||
self.log(section)
|
||||
cont_id = sec['key']
|
||||
|
||||
articles = []
|
||||
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_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))
|
||||
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']
|
||||
art_id = arts['id']
|
||||
self.log(' ', title, '\n\t', desc)
|
||||
art_cont = self.get_article(art_id)
|
||||
pt = PersistentTemporaryFile('.html')
|
||||
pt.write(art_cont)
|
||||
pt.close()
|
||||
url = 'file:///' + pt.name
|
||||
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 get_article(self, article_id):
|
||||
from calibre import browser
|
||||
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()
|
||||
|
||||
def populate_article_metadata(self, article, soup, first):
|
||||
lnk = soup.find('div', attrs={'id':'share-link'})
|
||||
lnk = soup.find('div', attrs={'id': 'share-link'})
|
||||
if lnk:
|
||||
article.url = lnk['title']
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user