mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-06-23 15:30:45 -04:00
Merge branch 'master' of https://github.com/unkn0w7n/calibre
This commit is contained in:
Kovid Goyal
commit
b21ae6f411
@ -283,7 +283,7 @@ class Economist(BasicNewsRecipe):
|
||||
def get_browser(self, *args, **kwargs):
|
||||
if self.from_archive:
|
||||
kwargs['user_agent'] = (
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Lamarr'
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Liskov'
|
||||
)
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
else:
|
||||
|
||||
@ -77,6 +77,6 @@ class Espresso(BasicNewsRecipe):
|
||||
|
||||
def get_browser(self, *args, **kwargs):
|
||||
kwargs['user_agent'] = (
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Lamarr'
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Liskov'
|
||||
)
|
||||
return BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
|
||||
@ -283,7 +283,7 @@ class Economist(BasicNewsRecipe):
|
||||
def get_browser(self, *args, **kwargs):
|
||||
if self.from_archive:
|
||||
kwargs['user_agent'] = (
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Lamarr'
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Liskov'
|
||||
)
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
else:
|
||||
|
||||
@ -10,38 +10,10 @@ from uuid import uuid4
|
||||
from html5_parser import parse
|
||||
from lxml import etree
|
||||
|
||||
from calibre import replace_entities
|
||||
from calibre.ebooks.BeautifulSoup import NavigableString, Tag
|
||||
from calibre.ptempfile import PersistentTemporaryFile
|
||||
from calibre.web.feeds.news import BasicNewsRecipe
|
||||
|
||||
|
||||
def E(parent, name, text='', **attrs):
|
||||
ans = parent.makeelement(name, **attrs)
|
||||
ans.text = text
|
||||
parent.append(ans)
|
||||
return ans
|
||||
|
||||
|
||||
def process_node(node, html_parent):
|
||||
ntype = node.get('type')
|
||||
if ntype == 'tag':
|
||||
c = html_parent.makeelement(node['name'])
|
||||
c.attrib.update({k: v or '' for k, v in node.get('attribs', {}).items()})
|
||||
html_parent.append(c)
|
||||
for nc in node.get('children', ()):
|
||||
process_node(nc, c)
|
||||
elif ntype == 'text':
|
||||
text = node.get('data')
|
||||
if text:
|
||||
text = replace_entities(text)
|
||||
if len(html_parent):
|
||||
t = html_parent[-1]
|
||||
t.tail = (t.tail or '') + text
|
||||
else:
|
||||
html_parent.text = (html_parent.text or '') + text
|
||||
|
||||
|
||||
def safe_dict(data, *names):
|
||||
ans = data
|
||||
for x in names:
|
||||
@ -49,37 +21,86 @@ def safe_dict(data, *names):
|
||||
return ans
|
||||
|
||||
|
||||
class JSONHasNoContent(ValueError):
|
||||
pass
|
||||
def process_web_list(li_node):
|
||||
li_html = ''
|
||||
for li in li_node['items']:
|
||||
if li.get('textHtml'):
|
||||
li_html += f'<li>{li.get("textHtml")}</li>'
|
||||
else:
|
||||
li_html += f'<li>{li.get("text", "")}</li>'
|
||||
return li_html
|
||||
|
||||
|
||||
def load_article_from_json(raw, root):
|
||||
def process_info_box(bx):
|
||||
info = ''
|
||||
for x in safe_dict(bx, 'components'):
|
||||
info += f'<blockquote>{process_web_node(x)}</blockquote>'
|
||||
return info
|
||||
|
||||
|
||||
def process_web_node(node):
|
||||
ntype = node.get('type', '')
|
||||
if ntype == 'CROSSHEAD':
|
||||
if node.get('textHtml'):
|
||||
return f'<h4>{node.get("textHtml")}</h4>'
|
||||
return f'<h4>{node.get("text", "")}</h4>'
|
||||
elif ntype in ['PARAGRAPH', 'BOOK_INFO']:
|
||||
if node.get('textHtml'):
|
||||
return f'<p>{node.get("textHtml")}</p>'
|
||||
return f'<p>{node.get("text", "")}</p>'
|
||||
elif ntype == 'IMAGE':
|
||||
alt = '' if node.get('altText') is None else node.get('altText')
|
||||
cap = ''
|
||||
if node.get('caption'):
|
||||
if node['caption'].get('textHtml') is not None:
|
||||
cap = node['caption']['textHtml']
|
||||
return f'<div><img src="{node["url"]}" title="{alt}"></div><div style="text-align:center; font-size:small;">{cap}</div>'
|
||||
elif ntype == 'PULL_QUOTE':
|
||||
if node.get('textHtml'):
|
||||
return f'<blockquote>{node.get("textHtml")}</blockquote>'
|
||||
return f'<blockquote>{node.get("text", "")}</blockquote>'
|
||||
elif ntype == 'DIVIDER':
|
||||
return '<hr>'
|
||||
elif ntype == 'INFOGRAPHIC':
|
||||
if node.get('fallback'):
|
||||
return process_web_node(node['fallback'])
|
||||
elif ntype == 'INFOBOX':
|
||||
return process_info_box(node)
|
||||
elif ntype == 'UNORDERED_LIST':
|
||||
if node.get('items'):
|
||||
return process_web_list(node)
|
||||
elif ntype:
|
||||
print('** ', ntype)
|
||||
return ''
|
||||
|
||||
|
||||
def load_article_from_web_json(raw):
|
||||
# open('/t/raw.json', 'w').write(raw)
|
||||
data = json.loads(raw)
|
||||
body = root.xpath('//body')[0]
|
||||
article = E(body, 'article')
|
||||
E(article, 'div', data['flyTitle'], style='color: red; font-size:small; font-weight:bold;')
|
||||
E(article, 'h1', data['title'], title=safe_dict(data, 'url', 'canonical') or '')
|
||||
E(article, 'div', data['rubric'], style='font-style: italic; color:#202020;')
|
||||
body = ''
|
||||
try:
|
||||
data = json.loads(raw)['props']['pageProps']['cp2Content']
|
||||
except Exception:
|
||||
data = json.loads(raw)['props']['pageProps']['content']
|
||||
body += f'<div style="color: red; font-size:small; font-weight:bold;">{data.get("flyTitle", "")}</div>'
|
||||
body += f'<h1>{data["headline"]}</h1>'
|
||||
if data.get('rubric') and data.get('rubric') is not None:
|
||||
body += f'<div style="font-style: italic; color:#202020;">{data.get("rubric", "")}</div>'
|
||||
try:
|
||||
date = data['dateModified']
|
||||
except Exception:
|
||||
date = data['datePublished']
|
||||
dt = datetime.fromisoformat(date[:-1]) + timedelta(seconds=time.timezone)
|
||||
dt = dt.strftime('%b %d, %Y %I:%M %p')
|
||||
if data['dateline'] is None:
|
||||
E(article, 'p', dt, style='color: gray; font-size:small;')
|
||||
if data.get('dateline') is None:
|
||||
body += f'<p style="color: gray; font-size: small;">{dt}</p>'
|
||||
else:
|
||||
E(article, 'p', dt + ' | ' + (data['dateline']), style='color: gray; font-size:small;')
|
||||
main_image_url = safe_dict(data, 'image', 'main', 'url').get('canonical')
|
||||
body += f'<p style="color: gray; font-size: small;">{dt + " | " + (data["dateline"])}</p>'
|
||||
main_image_url = safe_dict(data, 'leadComponent') or ''
|
||||
if main_image_url:
|
||||
div = E(article, 'div')
|
||||
try:
|
||||
E(div, 'img', src=main_image_url)
|
||||
except Exception:
|
||||
pass
|
||||
for node in data.get('text') or ():
|
||||
process_node(node, article)
|
||||
body += process_web_node(data['leadComponent'])
|
||||
for node in data.get('body'):
|
||||
body += process_web_node(node)
|
||||
return '<html><body><article>' + body + '</article></body></html>'
|
||||
|
||||
|
||||
def cleanup_html_article(root):
|
||||
@ -172,6 +193,8 @@ class EconomistNews(BasicNewsRecipe):
|
||||
|
||||
needs_subscription = False
|
||||
|
||||
from_web = False
|
||||
|
||||
recipe_specific_options = {
|
||||
'days': {
|
||||
'short': 'Oldest article to download from this news source. In days ',
|
||||
@ -192,16 +215,25 @@ class EconomistNews(BasicNewsRecipe):
|
||||
self.oldest_article = float(d)
|
||||
|
||||
def get_browser(self, *args, **kwargs):
|
||||
kwargs['user_agent'] = 'TheEconomist-Lamarr-android'
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
br.addheaders += [
|
||||
('accept', '*/*'),
|
||||
('content-type', 'application/json'),
|
||||
('apollographql-client-name', 'mobile-app-apollo'),
|
||||
('apollographql-client-version', '3.50.0'),
|
||||
('x-request-id', str(uuid4())),
|
||||
]
|
||||
return br
|
||||
if self.from_web:
|
||||
kwargs['user_agent'] = (
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Liskov'
|
||||
)
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
else:
|
||||
kwargs['user_agent'] = 'TheEconomist-Liskov-android'
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
br.addheaders += [
|
||||
('accept', 'multipart/mixed; deferSpec=20220824, application/json'),
|
||||
('accept-encoding', 'gzip'),
|
||||
('content-type', 'application/json'),
|
||||
('x-app-trace-id', str(uuid4())),
|
||||
('x-economist-consumer', 'TheEconomist-Liskov-android'),
|
||||
('x-teg-client-name', 'Economist-Android'),
|
||||
('x-teg-client-os', 'Android'),
|
||||
('x-teg-client-version', '4.30.0'),
|
||||
]
|
||||
return br
|
||||
|
||||
def economist_return_index(self, ans):
|
||||
if not ans:
|
||||
@ -215,9 +247,9 @@ class EconomistNews(BasicNewsRecipe):
|
||||
|
||||
def parse_index(self):
|
||||
query = {
|
||||
'query': 'query HomeQuery($homeId:String!$relatedId:String!){canonical(ref:$homeId){hasPart{parts{id title:headline cta{text url __typename}type hasPart{parts{...ArticleFragment ...VideoFragment hasPart{parts{...ArticleFragment isPartOf{id context{title:headline flyTitle:subheadline rubric:description dateline image{...ImageMainFragment ...ImagePromoFragment __typename}__typename}__typename}__typename}__typename}isPartOf{id context{title:headline flyTitle:subheadline rubric:description dateline image{...ImageMainFragment ...ImagePromoFragment __typename}__typename}__typename}__typename}__typename}__typename}__typename}__typename}relatedTopStories:canonical(ref:$relatedId){id title:headline hasPart(size:2 sort:"datePublished:desc"){parts{...ArticleFragment isPartOf{id context{title:headline flyTitle:subheadline rubric:description dateline image{...ImageMainFragment ...ImagePromoFragment __typename}__typename}__typename}__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 VideoFragment on Content{video{playlist{playlistId __typename}__typename}__typename}', # noqa: E501
|
||||
'operationName': 'HomeQuery',
|
||||
'variables': '{"homeId":"/content/mgo2tcc3u3002m4gndvffg3kqv7n5n3g","relatedId":"/content/bp252fp9p9dvkn6pcjog5cks9hhnrf96"}',
|
||||
'operationName': 'FindHomepage',
|
||||
'variables': '{"homepageType":"MOBILE"}',
|
||||
'query': 'query FindHomepage($homepageType: HomepageType!) { findHomepage(homepageType: $homepageType) { __typename ...HomepageFragment } } fragment CtaFragment on Cta { link text } fragment ContentIdentityFragment on ContentIdentity { articleType forceAppWebView leadMediaType } fragment ImageTeaserFragment on ImageComponent { altText height imageType source url width } fragment NarrationFragment on Narration { album bitrate duration filename id provider url } fragment PodcastAudioFragment on PodcastEpisode { id audio { url durationInSeconds } } fragment ArticleTeaserFragment on Article { id tegId url rubric headline flyTitle brand byline dateFirstPublished dateline dateModified datePublished dateRevised estimatedReadTime wordCount printHeadline contentIdentity { __typename ...ContentIdentityFragment } section { tegId name } teaserImage { __typename type ...ImageTeaserFragment } leadComponent { __typename type ...ImageTeaserFragment } narration { __typename ...NarrationFragment } podcast { __typename ...PodcastAudioFragment } } fragment OverridesFragment on Overrides { flyTitle headline rubric teaserImage { __typename ...ImageTeaserFragment } } fragment CollectionItemFragment on CollectionItem { __typename type ... on CollectionArticleItem { article { __typename ...ArticleTeaserFragment } overrides { __typename ...OverridesFragment } } ... on CollectionRelatedArticleItem { article { __typename ...ArticleTeaserFragment } overrides { __typename ...OverridesFragment } } ... on CollectionExternalLinkItem { url overrides { __typename ...OverridesFragment } } } fragment HomepageFragment on Homepage { components { __typename id headline type ... on StandardCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on TakeoverCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on DiscoverRailCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on TopStoriesCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on EmbedsCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on CarouselCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } imageLayout variant } ... on VideoCarouselCollection { cta { __typename ...CtaFragment } playlistId source fallbackStoryLink { sourceId } } ... on CoverPackageCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on LatestEditionCollection { cta { __typename ...CtaFragment } items { __typename ...CollectionItemFragment } } ... on Newsletter { slug items { __typename ...CollectionItemFragment } } } }', # noqa: E501
|
||||
}
|
||||
url = 'https://cp2-graphql-gateway.p.aws.economist.com/graphql?' + urlencode(query, safe='()!', quote_via=quote)
|
||||
try:
|
||||
@ -228,41 +260,39 @@ class EconomistNews(BasicNewsRecipe):
|
||||
return self.economist_return_index(ans)
|
||||
|
||||
def economist_parse_index(self, raw):
|
||||
data = json.loads(raw)['data']['canonical']['hasPart']['parts']
|
||||
data = json.loads(raw)['data']['findHomepage']['components']
|
||||
|
||||
feeds = []
|
||||
|
||||
for part in data:
|
||||
section = part.get('title', 'Articles')
|
||||
if not part.get('items'):
|
||||
continue
|
||||
section = part.get('headline', 'Articles')
|
||||
self.log(section)
|
||||
|
||||
articles = []
|
||||
|
||||
for art in part['hasPart']['parts']:
|
||||
title = safe_dict(art, 'title')
|
||||
desc = safe_dict(art, 'rubric') or ''
|
||||
sub = safe_dict(art, 'flyTitle') or ''
|
||||
for art in part['items']:
|
||||
if not art.get('article'):
|
||||
continue
|
||||
arts = art['article']
|
||||
title = safe_dict(arts, 'headline')
|
||||
desc = safe_dict(arts, 'rubric') or ''
|
||||
sub = safe_dict(arts, 'flyTitle') or ''
|
||||
if sub and section != sub:
|
||||
desc = sub + ' :: ' + desc
|
||||
if not art.get('text'):
|
||||
continue
|
||||
date_ = art['datePublished']
|
||||
date_ = arts['datePublished']
|
||||
dt = datetime.fromisoformat(date_[:-1]) + timedelta(seconds=time.timezone)
|
||||
if (datetime.now() - dt) > timedelta(self.oldest_article):
|
||||
continue
|
||||
pt = PersistentTemporaryFile('.html')
|
||||
pt.write(json.dumps(art).encode('utf-8'))
|
||||
pt.close()
|
||||
url = 'file:///' + pt.name
|
||||
url = process_url(arts['url'])
|
||||
articles.append({'title': title, 'url': url, 'description': desc})
|
||||
self.log('\t', title, '\n\t\t', desc)
|
||||
if articles:
|
||||
feeds.append((section, articles))
|
||||
self.from_web = True
|
||||
return feeds
|
||||
|
||||
def populate_article_metadata(self, article, soup, first):
|
||||
article.url = soup.find('h1')['title']
|
||||
|
||||
def preprocess_html(self, soup):
|
||||
width = '600'
|
||||
w = self.recipe_specific_options.get('res')
|
||||
@ -275,16 +305,17 @@ class EconomistNews(BasicNewsRecipe):
|
||||
|
||||
def preprocess_raw_html(self, raw, url):
|
||||
# open('/t/raw.html', 'wb').write(raw.encode('utf-8'))
|
||||
|
||||
body = '<html><body><article></article></body></html>'
|
||||
root = parse(body)
|
||||
load_article_from_json(raw, root)
|
||||
|
||||
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.'
|
||||
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__"]')
|
||||
|
||||
html = load_article_from_web_json(script[0].text)
|
||||
|
||||
root = parse(html)
|
||||
for div in root.xpath('//div[@class="lazy-image"]'):
|
||||
noscript = list(div.iter('noscript'))
|
||||
if noscript and noscript[0].text:
|
||||
|
||||
@ -185,7 +185,7 @@ class econ_search(BasicNewsRecipe):
|
||||
|
||||
def get_browser(self, *args, **kwargs):
|
||||
kwargs['user_agent'] = (
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Lamarr'
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Liskov'
|
||||
)
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
return br
|
||||
|
||||
@ -203,7 +203,7 @@ class EconomistWorld(BasicNewsRecipe):
|
||||
|
||||
def get_browser(self, *args, **kwargs):
|
||||
kwargs['user_agent'] = (
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Lamarr'
|
||||
'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.6533.103 Mobile Safari/537.36 Liskov'
|
||||
)
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
return br
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user