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

Allow nyt todays paper download to work even when page errors out

Browse Source
This commit is contained in:
Kovid Goyal 2018-11-05 08:36:32 +05:30
parent 3897ade179
commit 661c47501a
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 14 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
recipes/nytimes_sub.recipe
View File

@ -140,7 +140,20 @@ class NewYorkTimes(BasicNewsRecipe):
def read_nyt_metadata(self):
INDEX = 'https://www.nytimes.com/section/todayspaper'
# INDEX = 'file:///t/raw.html'
soup = self.index_to_soup(INDEX)
try:
soup = self.index_to_soup(INDEX)
except Exception as err:
if getattr(err, 'code', None) == 404:
try:
soup = self.index_to_soup(strftime('https://www.nytimes.com/issue/todayspaper/%Y/%m/%d/todays-new-york-times'))
except Exception as err:
if getattr(err, 'code', None) == 404:
dt = datetime.datetime.today() - datetime.timedelta(days=1)
soup = self.index_to_soup(dt.strftime('https://www.nytimes.com/issue/todayspaper/%Y/%m/%d/todays-new-york-times'))
else:
raise
else:
raise
pdate = soup.find('meta', attrs={'name':'pdate', 'content': True})['content']
date = strptime(pdate, '%Y%m%d', assume_utc=False, as_utc=False)
self.cover_url = 'https://static01.nyt.com/images/{}/nytfrontpage/scan.jpg'.format(date.strftime('%Y/%m/%d'))