Backup_Repos/calibre
1
0
Fork 0
mirror of git://github.com/kovidgoyal/calibre.git synced 2025-12-06 10:05:38 +01:00
Code Issues Projects Releases Wiki Activity

workaround for nyt todays paper page using undefinedinstead of null in its json

Browse source
This commit is contained in:
Kovid Goyal 2022-02-18 19:59:22 +05:30
parent 803da449ba
commit 07b4239634
No known key found for this signature in database
GPG key ID: 06BC317B515ACE7C
2 changed files with 4 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
recipes/nytimes.recipe
View file

@ -186,7 +186,8 @@ def parse_todays_page(self):
soup = self.read_nyt_metadata() soup = self.read_nyt_metadata()
script = soup.findAll('script', text=lambda x: x and 'window.__preloadedData' in x)[0] script = soup.findAll('script', text=lambda x: x and 'window.__preloadedData' in x)[0]
script = type(u'')(script) script = type(u'')(script)
data = json.loads(script[script.find('{'):script.rfind(';')].strip().rstrip(';'))['initialState'] json_data = script[script.find('{'):script.rfind(';')].strip().rstrip(';')
data = json.loads(json_data.replace(':undefined', ':null'))['initialState']
containers, sections = {}, {} containers, sections = {}, {}
article_map = {} article_map = {}
gc_pat = re.compile(r'groupings.(\d+).containers.(\d+)') gc_pat = re.compile(r'groupings.(\d+).containers.(\d+)')

3
recipes/nytimes_sub.recipe
View file

@ -186,7 +186,8 @@ def parse_todays_page(self):
soup = self.read_nyt_metadata() soup = self.read_nyt_metadata()
script = soup.findAll('script', text=lambda x: x and 'window.__preloadedData' in x)[0] script = soup.findAll('script', text=lambda x: x and 'window.__preloadedData' in x)[0]
script = type(u'')(script) script = type(u'')(script)
data = json.loads(script[script.find('{'):script.rfind(';')].strip().rstrip(';'))['initialState'] json_data = script[script.find('{'):script.rfind(';')].strip().rstrip(';')
data = json.loads(json_data.replace(':undefined', ':null'))['initialState']
containers, sections = {}, {} containers, sections = {}, {}
article_map = {} article_map = {}
gc_pat = re.compile(r'groupings.(\d+).containers.(\d+)') gc_pat = re.compile(r'groupings.(\d+).containers.(\d+)')