diff --git a/recipes/bloomberg-business-week.recipe b/recipes/bloomberg-business-week.recipe
index 941604dfb0..8b241b1dee 100644
--- a/recipes/bloomberg-business-week.recipe
+++ b/recipes/bloomberg-business-week.recipe
@@ -1,5 +1,6 @@
from calibre.web.feeds.news import BasicNewsRecipe, classes
from calibre import browser
+from html5_parser import parse
import json
import re
@@ -19,7 +20,7 @@ class Bloomberg(BasicNewsRecipe):
#auth {font-size:small; font-weight:bold;}
#time, .chart {font-size:small;}
#subhead {font-style:italic; color:#404040;}
- .news-figure-caption-text, #cap {font-size:small; text-align:center;}
+ .news-figure-caption-text, #cap, #img {font-size:small; text-align:center;}
.news-figure-credit {font-size:small; text-align:center; color:#202020;}
'''
@@ -70,17 +71,22 @@ class Bloomberg(BasicNewsRecipe):
return feeds
def preprocess_raw_html(self, raw, *a):
- m = re.search('data-component-props="ArticleBody">', raw)
+ root = parse(raw)
+ m = root.xpath('//script[@data-component-props="ArticleBody"]')
if not m:
- m = re.search('data-component-props="FeatureBody">', raw)
+ m = root.xpath('//script[@data-component-props="FeatureBody"]')
if not m:
- return raw
-
- raw = raw[m.start():]
- raw = raw.split('>', 1)[1]
- data = json.JSONDecoder().raw_decode(raw)[0]
- data = data['story']
+ m2 = root.xpath('//script[@id="__NEXT_DATA__"]')
+ if not m2:
+ return raw
+ if m:
+ data = json.loads(m[0].text)
+ data = data['story']
+ else:
+ data = json.loads(m2[0].text)
+ data = data['props']['pageProps']['story']
+
title = '' + data['headline'] + '
'
cat = subhead = lede = auth = caption = ''
@@ -104,11 +110,39 @@ class Bloomberg(BasicNewsRecipe):
if 'ledeImageUrl' in data:
if data['ledeImageUrl'] is not None:
lede = '
'.format(data['ledeImageUrl'])
+
+ if 'ledeDescription' in data:
+ if data['ledeDescription'] is not None:
+ caption = '' + data['ledeDescription'] + ''
+ else:
+ if 'lede' in data:
+ if data['lede'] is not None:
+ if 'alt' in data['lede']:
+ if data['lede']['alt'] is not None:
+ caption = '' + data['lede']['alt'] + ''
- if data['ledeDescription'] is not None:
- caption = '' + data['ledeDescription'] + ''
+ if m:
+ body = data['body']
+ else:
+ body = ''
+ body_data = data['body']['content']
+ for objects in body_data:
+ if objects['type'] == 'media':
+ if objects['subType'] == 'photo':
+ body += '
'.format(objects['data']['photo']['src'])
+ body += '' + objects['data']['photo']['caption'] + '
'
+ if objects['type'] == 'paragraph' or 'heading':
+ body += ''
+ if not 'content' in objects:
+ continue
+ for content in objects['content']:
+ if 'value' in content:
+ body += content['value']
+ elif 'content' in content:
+ for val_cont in content['content']:
+ if 'value' in val_cont:
+ body += val_cont['value']
- body = data['body']
html = '
' + cat + title + subhead + auth + lede + caption + '' + body
return html