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

Update New York Times Book Review

Browse Source 
They rationalised their JSON schema finally.
This commit is contained in:
Kovid Goyal 2022-09-11 18:55:32 +05:30
parent 3d82d3f70d
commit b8b4351741
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 31 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

66
recipes/nytimesbook.recipe
View File

@ -12,17 +12,11 @@ from calibre.web.feeds.news import BasicNewsRecipe
# {{{ parse NYT JSON
def key_startswith(key, obj):
for q, val in obj.items():
if q.startswith(key):
return val
def is_heading(tn):
return tn in ('Heading1Block', 'Heading2Block', 'Heading3Block', 'Heading4Block')
def process_inline_text(lines, block, data):
def process_inline_text(lines, block):
text = ''
if 'text@stripHtml' in block:
text = escape(block['text@stripHtml'])
@ -32,16 +26,16 @@ def process_inline_text(lines, block, data):
text = block['text']
if text:
for fmt in block.get('formats', ()):
tn = fmt['typename']
tn = fmt['__typename']
if tn == 'LinkFormat':
ab = data[fmt['id']]
ab = fmt
text = '<a href="{}" title="{}">{}</a>'.format(ab['url'], ab.get('title') or '', text)
elif tn == 'BoldFormat':
text = '<b>' + text + '</b>'
lines.append(text)
def process_paragraph(lines, block, data, content_key='content'):
def process_paragraph(lines, block, content_key='content'):
tn = block['__typename']
m = re.match(r'Heading([1-6])Block', tn)
if m is not None:
@ -52,48 +46,48 @@ def process_paragraph(lines, block, data, content_key='content'):
style = 'text-align: {}'.format(ta.lower())
lines.append('<{} style="{}">'.format(tag, style))
for item in block[content_key]:
tn = item['typename']
tn = item['__typename']
if tn in ('TextInline', 'Byline'):
process_inline_text(lines, data[item['id']], data)
process_inline_text(lines, item)
lines.append('</' + tag + '>')
def process_timestamp(lines, block, data):
def process_timestamp(lines, block):
ts = block['timestamp']
dt = parse_iso8601(ts, as_utc=False)
lines.append('<p class="timestamp">' + escape(dt.strftime('%b %d, %Y')) + '</p>')
def process_header(lines, block, data):
def process_header(lines, block):
label = block.get('label')
if label:
process_paragraph(lines, data[label['id']], data)
process_paragraph(lines, label)
headline = block.get('headline')
if headline:
process_paragraph(lines, data[headline['id']], data)
process_paragraph(lines, headline)
summary = block.get('summary')
if summary:
process_paragraph(lines, data[summary['id']], data)
process_paragraph(lines, summary)
lm = block.get('ledeMedia')
if lm and lm.get('typename') == 'ImageBlock':
process_image_block(lines, data[lm['id']], data)
if lm and lm.get('__typename') == 'ImageBlock':
process_image_block(lines, lm)
byline = block.get('byline')
if byline:
process_paragraph(lines, data[byline['id']], data, content_key='bylines')
process_paragraph(lines, byline, content_key='bylines')
timestamp = block.get('timestampBlock')
if timestamp:
process_timestamp(lines, data[timestamp['id']], data)
process_timestamp(lines, timestamp)
def process_image_block(lines, block, data):
media = data[block['media']['id']]
def process_image_block(lines, block):
media = block['media']
caption = media.get('caption')
caption_lines = []
if caption:
process_inline_text(caption_lines, data[caption['id']], data)
crops = key_startswith('crops({', media)
renditions = data[crops[0]['id']]['renditions']
img = data[renditions[0]['id']]['url']
process_inline_text(caption_lines, caption)
crops = media['crops']
renditions = crops[0]['renditions']
img = renditions[0]['url']
lines.append('<div style="text-align: center"><img src={}/>'.format(quoteattr(img)))
lines.extend(caption_lines)
lines.append('</div>')
@ -101,18 +95,19 @@ def process_image_block(lines, block, data):
def json_to_html(raw):
data = json.loads(raw.replace(':undefined', ':null'))
data = data['initialState']
# open('/t/raw.json', 'w').write(json.dumps(data, indent=2))
data = data['initialData']['data']
article = next(iter(data.values()))
body = data[article['sprinkledBody']['id']]
body = article['sprinkledBody']['content']
lines = []
for item in body['content@filterEmpty']:
tn = item['typename']
if tn in ('HeaderBasicBlock', 'HeaderLegacyBlock'):
process_header(lines, data[item['id']], data)
for item in body:
tn = item['__typename']
if tn in ('HeaderBasicBlock', 'HeaderLegacyBlock', 'HeaderFullBleedVerticalBlock'):
process_header(lines, item)
elif tn in ('ParagraphBlock', 'LabelBlock', 'DetailBlock') or is_heading(tn):
process_paragraph(lines, data[item['id']], data)
process_paragraph(lines, item)
elif tn == 'ImageBlock':
process_image_block(lines, data[item['id']], data)
process_image_block(lines, item)
return '<html><body>' + '\n'.join(lines) + '</body></html>'
@ -152,6 +147,7 @@ class NewYorkTimesBookReview(BasicNewsRecipe):
return html
def parse_index(self):
return [('Articles', [{'url': 'https://www.nytimes.com/2022/09/08/books/review/karen-armstrong-by-the-book-interview.html', 'title':'test'}])]
soup = self.index_to_soup(
'https://www.nytimes.com/pages/books/review/index.html')