Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

...

Browse Source
This commit is contained in:
unkn0w7n 2024-11-28 10:23:58 +05:30
parent ea2d3f9edd
commit 0adbd16f21
1 changed files with 11 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
recipes/economist_world_ahead.recipe
View File

@ -49,6 +49,7 @@ def process_node(node):
print('** ', ntype) print('** ', ntype)
return '' return ''
def safe_dict(data, *names): def safe_dict(data, *names):
ans = data ans = data
for x in names: for x in names:
@ -187,15 +188,14 @@ class EconomistWorld(BasicNewsRecipe):
} }
def get_browser(self, *args, **kwargs): def get_browser(self, *args, **kwargs):
kwargs['user_agent'] = 'Mozilla/5.0 (Linux; Android 14; 330333QCG Build/AP1A.140705.005; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/125.0.6422.165 Mobile Safari/537.36 Lamarr/3.37.0-3037003 (android)' # noqa kwargs['user_agent'] = (
'Mozilla/5.0 (Linux; Android 14; 330333QCG Build/AP1A.140705.005; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/125.0.6422.165 Mobile Safari/537.36 Lamarr/3.37.0-3037003 (android)' # noqa
)
br = BasicNewsRecipe.get_browser(self, *args, **kwargs) br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
br.addheaders += [ br.addheaders += [('x-requested-with', 'com.economist.lamarr')]
('x-requested-with', 'com.economist.lamarr')
]
return br return br
def economist_test_article(self): def economist_test_article(self):
self.cover_url = None
return [('Articles', [{'title':'test', return [('Articles', [{'title':'test',
'url':'https://www.economist.com/the-world-ahead/2024/11/20/ten-business-trends-for-2025-and-forecasts-for-15-industries' 'url':'https://www.economist.com/the-world-ahead/2024/11/20/ten-business-trends-for-2025-and-forecasts-for-15-industries'
}])] }])]
@ -257,9 +257,12 @@ class EconomistWorld(BasicNewsRecipe):
# open('/t/raw.html', 'wb').write(raw.encode('utf-8')) # open('/t/raw.html', 'wb').write(raw.encode('utf-8'))
root_ = parse(raw) root_ = parse(raw)
if '/interactive/' in url: if '/interactive/' in url:
return '<html><body><article><h1>' + root_.xpath('//h1')[0].text + '</h1><em>' \ return (
+ 'This article is supposed to be read in a browser' \ '<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>' + '</em></article></body></html>'
)
script = root_.xpath('//script[@id="__NEXT_DATA__"]') script = root_.xpath('//script[@id="__NEXT_DATA__"]')