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

Merge branch 'ByteTheBubble-fix-json-parsing-nyt' of https://github.com/ByteTheBubble/calibre

Browse Source
This commit is contained in:
Kovid Goyal 2025-05-30 17:06:31 +05:30
commit 4fd2a30f78
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
2 changed files with 34 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
recipes/nytimes.recipe
View File

@ -82,6 +82,17 @@ def new_tag(soup, name, attrs=()):
return Tag(soup, name, attrs=attrs or None) return Tag(soup, name, attrs=attrs or None)
def clean_js_json(text):
text = text.replace('undefined', 'null')
text = re.sub(
r',?\s*"[^"]+"\s*:\s*function\s*\([^)]*\)\s*\{.*?\}',
'',
text,
flags=re.DOTALL
)
return text
class NewYorkTimes(BasicNewsRecipe): class NewYorkTimes(BasicNewsRecipe):
if is_web_edition: if is_web_edition:
title = 'The New York Times (Web)' title = 'The New York Times (Web)'
@ -127,7 +138,8 @@ class NewYorkTimes(BasicNewsRecipe):
return self.index_to_soup(url, raw=True) return self.index_to_soup(url, raw=True)
def preprocess_raw_html(self, raw_html, url): def preprocess_raw_html(self, raw_html, url):
return self.nyt_parser.extract_html(self.index_to_soup(raw_html), url) cleaned = clean_js_json(raw_html)
return self.nyt_parser.extract_html(self.index_to_soup(cleaned), url)
articles_are_obfuscated = use_wayback_machine articles_are_obfuscated = use_wayback_machine
@ -212,8 +224,9 @@ class NewYorkTimes(BasicNewsRecipe):
self.nytimes_publication_date = pdate self.nytimes_publication_date = pdate
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)
json_data = script[script.find('{'):script.rfind(';')].strip().rstrip(';') # }} raw_json = script[script.find('{'):script.rfind(';')].strip().rstrip(';') # }}
self.nytimes_graphql_config = json.loads(json_data.replace(':undefined', ':null'))['config'] clean_json = clean_js_json(raw_json)
self.nytimes_graphql_config = json.loads(clean_json)['config']
return soup return soup
def nyt_graphql_query(self, qid, operationName='CollectionsQuery'): def nyt_graphql_query(self, qid, operationName='CollectionsQuery'):

21
recipes/nytimes_sub.recipe
View File

@ -82,6 +82,19 @@ def new_tag(soup, name, attrs=()):
return Tag(soup, name, attrs=attrs or None) return Tag(soup, name, attrs=attrs or None)
def clean_js_json(text):
text = text.replace('undefined', 'null')
# drop any JS function definitions
text = re.sub(
r',?\s*"[^"]+"\s*:\s*function\s*\([^)]*\)\s*\{.*?\}',
'',
text,
flags=re.DOTALL
)
return text
class NewYorkTimes(BasicNewsRecipe): class NewYorkTimes(BasicNewsRecipe):
if is_web_edition: if is_web_edition:
title = 'The New York Times (Web)' title = 'The New York Times (Web)'
@ -127,7 +140,8 @@ class NewYorkTimes(BasicNewsRecipe):
return self.index_to_soup(url, raw=True) return self.index_to_soup(url, raw=True)
def preprocess_raw_html(self, raw_html, url): def preprocess_raw_html(self, raw_html, url):
return self.nyt_parser.extract_html(self.index_to_soup(raw_html), url) cleaned = clean_js_json(raw_html)
return self.nyt_parser.extract_html(self.index_to_soup(cleaned), url)
articles_are_obfuscated = use_wayback_machine articles_are_obfuscated = use_wayback_machine
@ -212,8 +226,9 @@ class NewYorkTimes(BasicNewsRecipe):
self.nytimes_publication_date = pdate self.nytimes_publication_date = pdate
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)
json_data = script[script.find('{'):script.rfind(';')].strip().rstrip(';') # }} raw_json = script[script.find('{'):script.rfind(';')].strip().rstrip(';') # }}
self.nytimes_graphql_config = json.loads(json_data.replace(':undefined', ':null'))['config'] clean_json = clean_js_json(raw_json)
self.nytimes_graphql_config = json.loads(clean_json)['config']
return soup return soup
def nyt_graphql_query(self, qid, operationName='CollectionsQuery'): def nyt_graphql_query(self, qid, operationName='CollectionsQuery'):