mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
Update economist_free.recipe
This commit is contained in:
unkn0w7n
parent
4f5b833a78
commit
3b0fcaf924
@ -1,6 +1,7 @@
|
||||
#!/usr/bin/env python
|
||||
# License: GPLv3 Copyright: 2008, Kovid Goyal <kovid at kovidgoyal.net>
|
||||
|
||||
import re
|
||||
import json
|
||||
import time
|
||||
from collections import defaultdict
|
||||
@ -11,13 +12,12 @@ from html5_parser import parse
|
||||
from lxml import etree
|
||||
|
||||
from calibre import replace_entities
|
||||
from calibre.ebooks.BeautifulSoup import NavigableString, Tag
|
||||
from calibre.scraper.simple import read_url
|
||||
from calibre.ebooks.BeautifulSoup import BeautifulSoup, NavigableString, Tag
|
||||
from calibre.ptempfile import PersistentTemporaryFile
|
||||
from calibre.utils.date import parse_only_date
|
||||
from calibre.web.feeds.news import BasicNewsRecipe
|
||||
|
||||
use_archive = True
|
||||
|
||||
|
||||
def E(parent, name, text='', **attrs):
|
||||
ans = parent.makeelement(name, **attrs)
|
||||
@ -56,7 +56,6 @@ class JSONHasNoContent(ValueError):
|
||||
pass
|
||||
|
||||
|
||||
if use_archive:
|
||||
def load_article_from_json(raw, root):
|
||||
# open('/t/raw.json', 'w').write(raw)
|
||||
data = json.loads(raw)
|
||||
@ -84,8 +83,9 @@ if use_archive:
|
||||
pass
|
||||
for node in data.get('text') or ():
|
||||
process_node(node, article)
|
||||
else:
|
||||
def load_article_from_json(raw, root):
|
||||
|
||||
|
||||
def load_article_from_web_json(raw, root):
|
||||
# open('/t/raw.json', 'w').write(raw)
|
||||
try:
|
||||
data = json.loads(raw)['props']['pageProps']['content']
|
||||
@ -97,13 +97,13 @@ else:
|
||||
for child in tuple(body):
|
||||
body.remove(child)
|
||||
article = E(body, 'article')
|
||||
E(article, 'div', replace_entities(data['subheadline']) , style='color: red; font-size:small; font-weight:bold;')
|
||||
E(article, 'h1', replace_entities(data['headline']))
|
||||
E(article, 'div', replace_entities(data['description']), style='font-style: italic; color:#202020;')
|
||||
E(article, 'div', data['subheadline'], id='subhead')
|
||||
E(article, 'h1', data['headline'])
|
||||
E(article, 'p', data['description'], id='desc')
|
||||
if data['dateline'] is None:
|
||||
E(article, 'p', (data['datePublishedString'] or ''), style='color: gray; font-size:small;')
|
||||
E(article, 'p', (data['datePublishedString'] or ''), id='date')
|
||||
else:
|
||||
E(article, 'p', (data['datePublishedString'] or '') + ' | ' + (data['dateline']), style='color: gray; font-size:small;')
|
||||
E(article, 'p', (data['datePublishedString'] or '') + ' | ' + (data['dateline']), id='date')
|
||||
main_image_url = safe_dict(data, 'image', 'main', 'url').get('canonical')
|
||||
if main_image_url:
|
||||
div = E(article, 'div')
|
||||
@ -111,8 +111,7 @@ else:
|
||||
E(div, 'img', src=main_image_url)
|
||||
except Exception:
|
||||
pass
|
||||
for node in data.get('text') or ():
|
||||
process_node(node, article)
|
||||
E(article, 'section', id='body')
|
||||
|
||||
|
||||
def cleanup_html_article(root):
|
||||
@ -153,7 +152,6 @@ def process_url(url):
|
||||
|
||||
|
||||
class Economist(BasicNewsRecipe):
|
||||
|
||||
title = 'The Economist'
|
||||
language = 'en'
|
||||
encoding = 'utf-8'
|
||||
@ -167,12 +165,17 @@ class Economist(BasicNewsRecipe):
|
||||
extra_css = '''
|
||||
em { color:#202020; }
|
||||
img {display:block; margin:0 auto;}
|
||||
.sub { font-size:small; }
|
||||
#subhead { color: #404040; font-size:small; font-weight:bold; }'
|
||||
#desc { font-style: italic; color:#202020; }
|
||||
#date { color: gray; font-size:small; }
|
||||
'''
|
||||
oldest_article = 7.0
|
||||
resolve_internal_links = True
|
||||
remove_tags = [
|
||||
dict(name=['script', 'noscript', 'title', 'iframe', 'cf_floatingcontent', 'aside', 'footer']),
|
||||
dict(name=['script', 'noscript', 'title', 'iframe', 'cf_floatingcontent', 'aside', 'footer', 'svg']),
|
||||
dict(attrs={'aria-label': "Article Teaser"}),
|
||||
dict(attrs={'id': 'player'}),
|
||||
dict(attrs={
|
||||
'class': [
|
||||
'dblClkTrk', 'ec-article-info', 'share_inline_header',
|
||||
@ -186,6 +189,7 @@ class Economist(BasicNewsRecipe):
|
||||
),
|
||||
dict(attrs={
|
||||
'class': lambda x: x and 'blog-post__siblings-list-aside' in x.split()}),
|
||||
dict(attrs={'id': lambda x: x and 'gpt-ad-slot' in x}),
|
||||
classes(
|
||||
'share-links-header teaser--wrapped latest-updates-panel__container'
|
||||
' latest-updates-panel__article-link blog-post__section newsletter-form blog-post__bottom-panel'
|
||||
@ -198,6 +202,8 @@ class Economist(BasicNewsRecipe):
|
||||
# downloaded with connection reset by peer (104) errors.
|
||||
delay = 3
|
||||
|
||||
from_archive = False
|
||||
|
||||
recipe_specific_options = {
|
||||
'date': {
|
||||
'short': 'The date of the edition to download (YYYY-MM-DD format)',
|
||||
@ -245,7 +251,20 @@ class Economist(BasicNewsRecipe):
|
||||
)
|
||||
return ans
|
||||
|
||||
if use_archive:
|
||||
def get_content_id(self, ed_date):
|
||||
id_query = {
|
||||
'query': 'query EditionsQuery($from:Int$size:Int$ref:String!){section:canonical(ref:$ref){...EditionFragment __typename}}fragment EditionFragment on Content{hasPart(from:$from size:$size sort:"datePublished:desc"){total parts{id datePublished image{...ImageCoverFragment __typename}__typename}__typename}__typename}fragment ImageCoverFragment on Media{cover{headline width height url{canonical __typename}regionsAllowed __typename}__typename}',
|
||||
'operationName': 'EditionsQuery',
|
||||
'variables':'{"from":0,"size":24,"ref":"/content/d06tg8j85rifiq3oo544c6b9j61dno2n"}',
|
||||
}
|
||||
id_url = 'https://cp2-graphql-gateway.p.aws.economist.com/graphql?' + urlencode(id_query, safe='()!', quote_via=quote)
|
||||
raw_id_data = self.index_to_soup(id_url, raw=True)
|
||||
data = json.loads(raw_id_data)['data']['section']['hasPart']['parts']
|
||||
for x in data:
|
||||
if ed_date in x['datePublished']:
|
||||
return x['id']
|
||||
raise ValueError(ed_date, ' not found.')
|
||||
|
||||
def parse_index(self):
|
||||
edition_date = self.recipe_specific_options.get('date')
|
||||
# return self.economist_test_article()
|
||||
@ -256,23 +275,19 @@ class Economist(BasicNewsRecipe):
|
||||
'variables': '{"ref":"/content/d06tg8j85rifiq3oo544c6b9j61dno2n"}',
|
||||
}
|
||||
if edition_date and isinstance(edition_date, str):
|
||||
url = 'https://www.economist.com/weeklyedition/' + edition_date
|
||||
soup = self.index_to_soup(url)
|
||||
script_tag = soup.find("script", id="__NEXT_DATA__")
|
||||
if script_tag is None:
|
||||
raise ValueError('No script tag with JSON data found in the weeklyedition archive')
|
||||
data = json.loads(script_tag.string)
|
||||
content_id = data['props']['pageProps']['content']['id'].split('/')[-1]
|
||||
content_id = self.get_content_id(edition_date)
|
||||
query = {
|
||||
'query': 'query SpecificWeeklyEditionQuery($path:String!){section:canonical(ref:$path){...WeeklyEditionFragment __typename}}fragment WeeklyEditionFragment on Content{id type datePublished image{...ImageCoverFragment __typename}url{canonical __typename}hasPart(size:100 sort:"publication.context.position"){parts{...ArticleFragment __typename}__typename}__typename}fragment ArticleFragment on Content{ad{grapeshot{channels{name __typename}__typename}__typename}articleSection{internal{id title:headline __typename}__typename}audio{main{id duration(format:"seconds")source:channel{id __typename}url{canonical __typename}__typename}__typename}byline dateline dateModified datePublished dateRevised flyTitle:subheadline id image{...ImageInlineFragment ...ImageMainFragment ...ImagePromoFragment __typename}print{title:headline flyTitle:subheadline rubric:description section{id title:headline __typename}__typename}publication{id tegID title:headline flyTitle:subheadline datePublished regionsAllowed url{canonical __typename}__typename}rubric:description source:channel{id __typename}tegID text(format:"json")title:headline type url{canonical __typename}topic contentIdentity{forceAppWebview mediaType articleType __typename}__typename}fragment ImageInlineFragment on Media{inline{url{canonical __typename}width height __typename}__typename}fragment ImageMainFragment on Media{main{url{canonical __typename}width height __typename}__typename}fragment ImagePromoFragment on Media{promo{url{canonical __typename}id width height __typename}__typename}fragment ImageCoverFragment on Media{cover{headline width height url{canonical __typename}regionsAllowed __typename}__typename}', # noqa
|
||||
'query': 'query SpecificWeeklyEditionQuery($path:String!){section:canonical(ref:$path){...WeeklyEditionFragment __typename}}fragment WeeklyEditionFragment on Content{id type datePublished image{...ImageCoverFragment __typename}url{canonical __typename}hasPart(size:100 sort:"publication.context.position"){parts{...ArticleFragment __typename}__typename}__typename}fragment ArticleFragment on Content{articleSection{internal{id title:headline __typename}__typename}audio{main{id duration(format:"seconds")source:channel{id __typename}url{canonical __typename}__typename}__typename}byline dateline dateModified datePublished flyTitle:subheadline id image{...ImageInlineFragment ...ImageMainFragment ...ImagePromoFragment __typename}print{title:headline flyTitle:subheadline rubric:description section{id title:headline __typename}__typename}publication{id tegID title:headline flyTitle:subheadline datePublished regionsAllowed url{canonical __typename}__typename}rubric:description source:channel{id __typename}tegID text(format:"json")title:headline type url{canonical __typename}__typename}fragment ImageInlineFragment on Media{inline{url{canonical __typename}width height __typename}__typename}fragment ImageMainFragment on Media{main{url{canonical __typename}width height __typename}__typename}fragment ImagePromoFragment on Media{promo{url{canonical __typename}id width height __typename}__typename}fragment ImageCoverFragment on Media{cover{headline width height url{canonical __typename}regionsAllowed __typename}__typename}', # noqa
|
||||
'operationName': 'SpecificWeeklyEditionQuery',
|
||||
'variables': '{{"path":"/content/{}"}}'.format(content_id),
|
||||
'variables': '{{"path":"{}"}}'.format(content_id),
|
||||
}
|
||||
url = 'https://cp2-graphql-gateway.p.aws.economist.com/graphql?' + urlencode(query, safe='()!', quote_via=quote)
|
||||
try:
|
||||
raw = self.index_to_soup(url, raw=True)
|
||||
except Exception:
|
||||
raise ValueError('Server is not reachable, try again after some time.')
|
||||
self.log('Fetching articles from web archive.')
|
||||
self.from_archive = True
|
||||
return self.parse_web_index()
|
||||
ans = self.economist_parse_index(raw)
|
||||
return self.economist_return_index(ans)
|
||||
|
||||
@ -324,9 +339,12 @@ class Economist(BasicNewsRecipe):
|
||||
return [(section, articles) for section, articles in feeds_dict.items()]
|
||||
|
||||
def populate_article_metadata(self, article, soup, first):
|
||||
if not self.from_archive:
|
||||
article.url = soup.find('h1')['title']
|
||||
|
||||
def preprocess_html(self, soup):
|
||||
if self.from_archive:
|
||||
return self.preprocess_web_html(soup)
|
||||
width = '600'
|
||||
w = self.recipe_specific_options.get('res')
|
||||
if w and isinstance(w, str):
|
||||
@ -336,73 +354,15 @@ class Economist(BasicNewsRecipe):
|
||||
img['src'] = img['src'].replace('economist.com/', qua)
|
||||
return soup
|
||||
|
||||
else: # Load articles from individual article pages {{{
|
||||
|
||||
def __init__(self, *args, **kwargs):
|
||||
BasicNewsRecipe.__init__(self, *args, **kwargs)
|
||||
if self.output_profile.short_name.startswith('kindle'):
|
||||
# Reduce image sizes to get file size below amazon's email
|
||||
# sending threshold
|
||||
self.web2disk_options.compress_news_images = True
|
||||
self.web2disk_options.compress_news_images_auto_size = 5
|
||||
self.log.warn('Kindle Output profile being used, reducing image quality to keep file size below amazon email threshold')
|
||||
|
||||
def parse_index(self):
|
||||
edition_date = self.recipe_specific_options.get('date')
|
||||
# return self.economist_test_article()
|
||||
if edition_date and isinstance(edition_date, str):
|
||||
url = 'https://www.economist.com/weeklyedition/' + edition_date
|
||||
self.timefmt = ' [' + edition_date + ']'
|
||||
else:
|
||||
url = 'https://www.economist.com/weeklyedition'
|
||||
soup = self.index_to_soup(url)
|
||||
ans = self.economist_parse_index(soup)
|
||||
return self.economist_return_index(ans)
|
||||
|
||||
def economist_parse_index(self, soup):
|
||||
script_tag = soup.find("script", id="__NEXT_DATA__")
|
||||
if script_tag is not None:
|
||||
data = json.loads(script_tag.string)
|
||||
# open('/t/raw.json', 'w').write(json.dumps(data, indent=2, sort_keys=True))
|
||||
self.description = safe_dict(data, "props", "pageProps", "content", "image", "main", "headline")
|
||||
self.timefmt = ' [' + safe_dict(data, "props", "pageProps", "content", "datePublishedString") + ']'
|
||||
self.cover_url = safe_dict(data, "props", "pageProps", "content", "image", "main", "url", "canonical")
|
||||
self.log('Got cover:', self.cover_url)
|
||||
|
||||
feeds_dict = defaultdict(list)
|
||||
for part in safe_dict(data, "props", "pageProps", "content", "hasPart", "parts"):
|
||||
section = safe_dict(part, "print", "section", "headline") or ''
|
||||
title = safe_dict(part, "headline") or ''
|
||||
url = safe_dict(part, "url", "canonical") or ''
|
||||
if not section or not title or not url:
|
||||
continue
|
||||
desc = safe_dict(part, "description") or ''
|
||||
sub = safe_dict(part, "subheadline") or ''
|
||||
if sub and section != sub:
|
||||
desc = sub + ' :: ' + desc
|
||||
feeds_dict[section].append({"title": title, "url": url, "description": desc})
|
||||
self.log(' ', title, url, '\n ', desc)
|
||||
return [(section, articles) for section, articles in feeds_dict.items()]
|
||||
else:
|
||||
return []
|
||||
|
||||
# }}}
|
||||
|
||||
|
||||
def preprocess_raw_html(self, raw, url):
|
||||
if self.from_archive:
|
||||
return self.preprocess_raw_web_html(raw, url)
|
||||
|
||||
# open('/t/raw.html', 'wb').write(raw.encode('utf-8'))
|
||||
if use_archive:
|
||||
|
||||
body = '<html><body><article></article></body></html>'
|
||||
root = parse(body)
|
||||
load_article_from_json(raw, root)
|
||||
else:
|
||||
root = parse(raw)
|
||||
script = root.xpath('//script[@id="__NEXT_DATA__"]')
|
||||
if script:
|
||||
try:
|
||||
load_article_from_json(script[0].text, root)
|
||||
except JSONHasNoContent:
|
||||
cleanup_html_article(root)
|
||||
|
||||
if '/interactive/' in url:
|
||||
return '<html><body><article><h1>' + root.xpath('//h1')[0].text + '</h1><em>' \
|
||||
@ -495,11 +455,119 @@ class Economist(BasicNewsRecipe):
|
||||
url = url.rpartition('/')[0]
|
||||
return BasicNewsRecipe.canonicalize_internal_url(self, url, is_link=is_link)
|
||||
|
||||
# archive code
|
||||
def parse_web_index(self):
|
||||
edition_date = self.recipe_specific_options.get('date')
|
||||
# return self.economist_test_article()
|
||||
if edition_date and isinstance(edition_date, str):
|
||||
url = 'https://www.economist.com/weeklyedition/' + edition_date
|
||||
self.timefmt = ' [' + edition_date + ']'
|
||||
else:
|
||||
url = 'https://www.economist.com/weeklyedition'
|
||||
soup = self.index_to_soup(url)
|
||||
ans = self.economist_parse_web_index(soup)
|
||||
return self.economist_return_index(ans)
|
||||
|
||||
def get_login_cookies(username, password):
|
||||
print(33333333333, username, password)
|
||||
def economist_parse_web_index(self, soup):
|
||||
script_tag = soup.find("script", id="__NEXT_DATA__")
|
||||
if script_tag is not None:
|
||||
data = json.loads(script_tag.string)
|
||||
# open('/t/raw.json', 'w').write(json.dumps(data, indent=2, sort_keys=True))
|
||||
self.description = safe_dict(data, "props", "pageProps", "content", "image", "main", "headline")
|
||||
self.timefmt = ' [' + safe_dict(data, "props", "pageProps", "content", "formattedIssueDate") + ']'
|
||||
self.cover_url = safe_dict(data, "props", "pageProps", "content", "image", "main", "url", "canonical").replace(
|
||||
'economist.com/', 'economist.com/cdn-cgi/image/width=960,quality=80,format=auto/'
|
||||
)
|
||||
self.log('Got cover:', self.cover_url)
|
||||
|
||||
feeds_dict = defaultdict(list)
|
||||
for part in safe_dict(data, "props", "pageProps", "content", "hasPart", "parts"):
|
||||
section = safe_dict(part, "print", "section", "headline") or ''
|
||||
title = safe_dict(part, "headline") or ''
|
||||
url = safe_dict(part, "url", "canonical") or ''
|
||||
if not section or not title or not url:
|
||||
continue
|
||||
desc = safe_dict(part, "description") or ''
|
||||
sub = safe_dict(part, "subheadline") or ''
|
||||
if sub and section != sub:
|
||||
desc = sub + ' :: ' + desc
|
||||
feeds_dict[section].append({"title": title, "url": url, "description": desc})
|
||||
self.log(' ', title, url, '\n ', desc)
|
||||
return [(section, articles) for section, articles in feeds_dict.items()]
|
||||
else:
|
||||
return []
|
||||
|
||||
if __name__ == '__main__':
|
||||
import sys
|
||||
get_login_cookies(sys.argv[-2], sys.argv[-1])
|
||||
def preprocess_raw_web_html(self, raw, url):
|
||||
# open('/t/raw.html', 'wb').write(raw.encode('utf-8'))
|
||||
root = parse(raw)
|
||||
if '/interactive/' in url:
|
||||
return '<html><body><article><h1>' + root.xpath('//h1')[0].text + '</h1><em>' \
|
||||
+ 'This article is supposed to be read in a browser' \
|
||||
+ '</em></article></body></html>'
|
||||
script = root.xpath('//script[@id="__NEXT_DATA__"]')
|
||||
if script:
|
||||
try:
|
||||
load_article_from_web_json(script[0].text, root)
|
||||
except JSONHasNoContent:
|
||||
cleanup_html_article(root)
|
||||
for div in root.xpath('//div[@class="lazy-image"]'):
|
||||
noscript = list(div.iter('noscript'))
|
||||
if noscript and noscript[0].text:
|
||||
img = list(parse(noscript[0].text).iter('img'))
|
||||
if img:
|
||||
p = noscript[0].getparent()
|
||||
idx = p.index(noscript[0])
|
||||
p.insert(idx, p.makeelement('img', src=img[0].get('src')))
|
||||
p.remove(noscript[0])
|
||||
for x in root.xpath('//*[name()="script" or name()="style" or name()="source" or name()="meta"]'):
|
||||
x.getparent().remove(x)
|
||||
raw = etree.tostring(root, encoding='unicode')
|
||||
|
||||
raw_ar = read_url([], 'https://archive.is/latest/' + url)
|
||||
archive = BeautifulSoup(str(raw_ar))
|
||||
art = archive.find('article')
|
||||
if art:
|
||||
bdy = art.findAll('section')
|
||||
if len(bdy) != 0:
|
||||
content = bdy[-1]
|
||||
else:
|
||||
content = archive.find('div', attrs={'itemprop':'text'})
|
||||
soup = BeautifulSoup(raw)
|
||||
article = soup.find('section', attrs={'id':'body'})
|
||||
if not article:
|
||||
article = soup.find('div', attrs={'itemprop':'text'})
|
||||
if not article:
|
||||
article = soup.find(attrs={'itemprop':'blogPost'})
|
||||
if article and content:
|
||||
self.log('**fetching archive content')
|
||||
article.append(content)
|
||||
|
||||
div = soup.findAll(attrs={'style': lambda x: x and x.startswith(
|
||||
('color:rgb(13, 13, 13);', 'color: rgb(18, 18, 18);')
|
||||
)})
|
||||
for p in div:
|
||||
p.name = 'p'
|
||||
return str(soup)
|
||||
return raw
|
||||
return raw
|
||||
|
||||
def preprocess_web_html(self, soup):
|
||||
for img in soup.findAll('img', attrs={'old-src':True}):
|
||||
img['src'] = img['old-src']
|
||||
for a in soup.findAll('a', href=True):
|
||||
a['href'] = 'http' + a['href'].split('http')[-1]
|
||||
for fig in soup.findAll('figure'):
|
||||
fig['class'] = 'sub'
|
||||
for sty in soup.findAll(attrs={'style':True}):
|
||||
del sty['style']
|
||||
width = '600'
|
||||
w = self.recipe_specific_options.get('res')
|
||||
if w and isinstance(w, str):
|
||||
width = w
|
||||
for img in soup.findAll('img', src=True):
|
||||
if '/cdn-cgi/image/' not in img['src']:
|
||||
qua = 'economist.com/cdn-cgi/image/width=' + width + ',quality=80,format=auto/'
|
||||
img['src'] = img['src'].replace('economist.com/', qua)
|
||||
else:
|
||||
img['src'] = re.sub(r'width=\d+', 'width=' + width, img['src'])
|
||||
return soup
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user