mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-09-29 15:31:08 -04:00
243 lines
9.4 KiB
Python
243 lines
9.4 KiB
Python
#!/usr/bin/env python
|
|
# vim:fileencoding=utf-8
|
|
from __future__ import absolute_import, division, print_function, unicode_literals
|
|
|
|
import json
|
|
from datetime import date
|
|
from pprint import pformat
|
|
|
|
from calibre import prepare_string_for_xml as escape
|
|
from calibre.utils.iso8601 import parse_iso8601
|
|
from calibre.web.feeds.news import BasicNewsRecipe
|
|
|
|
|
|
def classes(classes):
|
|
q = frozenset(classes.split(' '))
|
|
return dict(attrs={
|
|
'class': lambda x: x and frozenset(x.split()).intersection(q)})
|
|
|
|
|
|
def extract_json(raw):
|
|
s = raw.find("window['__natgeo__']")
|
|
script = raw[s:raw.find('</script>', s)]
|
|
return json.loads(script[script.find('{'):].rstrip(';'))['page']['content']['prismarticle']
|
|
|
|
|
|
def parse_contributors(grp):
|
|
for item in grp:
|
|
line = '<div class="auth">' + escape(item['title']) + ' '
|
|
for c in item['contributors']:
|
|
line += escape(c['displayName'])
|
|
yield line + '</div>'
|
|
|
|
|
|
def parse_lead_image(media):
|
|
if 'image' in media:
|
|
yield '<p>'
|
|
if 'dsc' in media['image']:
|
|
yield '<div><img src="{}" alt="{}"></div>'.format(
|
|
escape(media['image']['src'], True), escape(media['image']['dsc'], True)
|
|
)
|
|
else:
|
|
yield '<div><img src="{}"></div>'.format(escape(media['image']['src'], True))
|
|
if 'caption' in media and 'credit' in media:
|
|
yield '<div class="cap">' + media['caption'] + '<span class="cred"> ' + media['credit'] + '</span></div>'
|
|
elif 'caption' in media:
|
|
yield '<div class="cap">' + media['caption'] + '</div>'
|
|
yield '</p>'
|
|
|
|
|
|
def parse_inline(inl):
|
|
if inl.get('content', {}).get('name', '') == 'Image':
|
|
props = inl['content']['props']
|
|
yield '<p>'
|
|
if 'image' in props:
|
|
yield '<div class="img"><img src="{}"></div>'.format(props['image']['src'])
|
|
if 'caption' in props:
|
|
yield '<div class="cap">{}<span class="cred">{}</span></div>'.format(
|
|
props['caption'].get('text', ''), ' ' + props['caption'].get('credit', '')
|
|
)
|
|
yield '</p>'
|
|
if inl.get('content', {}).get('name', '') == 'ImageGroup':
|
|
if 'images' in inl['content']['props']:
|
|
for imgs in inl['content']['props']['images']:
|
|
yield '<p>'
|
|
if 'src' in imgs:
|
|
yield '<div class="img"><img src="{}"></div>'.format(imgs['src'])
|
|
if 'caption' in imgs:
|
|
yield '<div class="cap">{}<span class="cred">{}</span></div>'.format(
|
|
imgs['caption'].get('text', ''), ' ' + imgs['caption'].get('credit', '')
|
|
)
|
|
yield '</p>'
|
|
|
|
|
|
def parse_cont(content):
|
|
for cont in content.get('content', {}):
|
|
if isinstance(cont, dict):
|
|
yield from parse_body(cont)
|
|
if isinstance(cont, str):
|
|
yield cont
|
|
|
|
|
|
def parse_body(x):
|
|
if isinstance(x, dict):
|
|
if 'type' in x:
|
|
tag = x['type']
|
|
if tag == 'inline':
|
|
yield ''.join(parse_inline(x))
|
|
elif 'attrs' in x and 'href' in x.get('attrs', ''):
|
|
yield '<' + tag + ' href="{}">'.format(x['attrs']['href'])
|
|
for yld in parse_cont(x):
|
|
yield yld
|
|
yield '</' + tag + '>'
|
|
else:
|
|
yield '<' + tag + '>'
|
|
for yld in parse_cont(x):
|
|
yield yld
|
|
yield '</' + tag + '>'
|
|
elif isinstance(x, list):
|
|
for y in x:
|
|
if isinstance(y, dict):
|
|
yield from parse_body(y)
|
|
|
|
|
|
def parse_article(edg):
|
|
sc = edg['schma']
|
|
yield '<div class="sub">' + escape(edg['sctn']) + '</div>'
|
|
yield '<h1>' + escape(sc['sclTtl']) + '</h1>'
|
|
yield '<div class="byline">' + escape(sc['sclDsc']) + '</div>'
|
|
yield '<p>'
|
|
for line in parse_contributors(edg.get('cntrbGrp', {})):
|
|
yield line
|
|
ts = parse_iso8601(edg['mdDt'], as_utc=False).strftime('%B %d, %Y')
|
|
yield '<div class="time">Published: ' + escape(ts) + '</div>'
|
|
if 'readTime' in edg:
|
|
yield '<div class="time">' + escape(edg['readTime']) + '</div>'
|
|
yield '</p>'
|
|
if edg.get('ldMda', {}).get('cmsType') == 'image':
|
|
for line in parse_lead_image(edg['ldMda']):
|
|
yield line
|
|
for main in edg['prismData']['mainComponents']:
|
|
if main['name'] == 'Body':
|
|
for item in main['props']['body']:
|
|
if isinstance(item, dict):
|
|
if item.get('type', '') == 'inline':
|
|
yield ''.join(parse_inline(item))
|
|
elif isinstance(item, list):
|
|
for line in item:
|
|
yield ''.join(parse_body(line))
|
|
|
|
|
|
def article_parse(data):
|
|
yield "<html><body>"
|
|
for frm in data['frms']:
|
|
if not frm:
|
|
continue
|
|
for mod in frm.get('mods', ()):
|
|
for edg in mod.get('edgs', ()):
|
|
if edg.get('cmsType') == 'ImmersiveLeadTile':
|
|
if 'image' in edg.get('cmsImage', {}):
|
|
for line in parse_lead_image(edg['cmsImage']):
|
|
yield line
|
|
if edg.get('cmsType') == 'ArticleBodyTile':
|
|
for line in parse_article(edg):
|
|
yield line
|
|
yield "</body></html>"
|
|
|
|
|
|
class NatGeo(BasicNewsRecipe):
|
|
title = u'National Geographic Magazine'
|
|
description = 'The National Geographic, an American monthly magazine'
|
|
language = 'en'
|
|
encoding = 'utf8'
|
|
publisher = 'nationalgeographic.com'
|
|
category = 'science, nat geo'
|
|
__author__ = 'Kovid Goyal, unkn0wn'
|
|
description = 'Inspiring people to care about the planet since 1888'
|
|
timefmt = ' [%a, %d %b, %Y]'
|
|
no_stylesheets = True
|
|
use_embedded_content = False
|
|
remove_attributes = ['style']
|
|
remove_javascript = False
|
|
masthead_url = 'https://i.natgeofe.com/n/e76f5368-6797-4794-b7f6-8d757c79ea5c/ng-logo-2fl.png?w=600&h=600'
|
|
remove_empty_feeds = True
|
|
resolve_internal_links = True
|
|
|
|
extra_css = '''
|
|
blockquote { color:#404040; }
|
|
.byline, i { font-style:italic; color:#202020; }
|
|
.cap { font-size:small; }
|
|
img {display:block; margin:0 auto;}
|
|
.cred { font-style:italic; font-size:small; color:#404040; }
|
|
.auth, .time, .sub { font-size:small; color:#5c5c5c; }
|
|
'''
|
|
|
|
recipe_specific_options = {
|
|
'date': {
|
|
'short': 'The date of the edition to download (Month-YYYY format)',
|
|
'long': 'For example, March-2023'
|
|
}
|
|
}
|
|
|
|
def parse_index(self):
|
|
edition = date.today().strftime('%B-%Y')
|
|
d = self.recipe_specific_options.get('date')
|
|
if d and isinstance(d, str):
|
|
edition = d
|
|
url = 'https://www.nationalgeographic.com/magazine/issue/' + edition.lower()
|
|
self.log('Downloading ', url)
|
|
self.timefmt = ' [' + edition + ']'
|
|
soup = self.index_to_soup(url)
|
|
# png = re.findall('https://i\.natgeofe\.com\S+?national-geographic-\S+?\.jpg', soup.decode('utf-8'))
|
|
# self.cover_url = png[0] + '?w=1000&h=1000'
|
|
self.cover_url = soup.find('meta', attrs={'property':'og:image'})['content'].split('?')[0] + '?w=1000'
|
|
|
|
# self.title = 'National Geographic ' + self.tag_to_string(name)
|
|
ans = {}
|
|
if photoart := soup.find(attrs={'class':lambda x: x and 'BgImagePromo__Container__Text__Link' in x.split()}):
|
|
section = 'Photo Essay'
|
|
title = self.tag_to_string(photoart)
|
|
url = photoart['href']
|
|
if url.startswith('/'):
|
|
url = 'https://www.nationalgeographic.com' + url
|
|
articles = ans.setdefault(section, [])
|
|
articles.append({'title': title, 'url': url})
|
|
for promo in soup.findAll(**classes('OneUpPromoCard__Content')):
|
|
url = promo.a['href']
|
|
section = self.tag_to_string(promo.find(**classes('SectionLabel')))
|
|
title = self.tag_to_string(promo.find(**classes('Card__Content__Heading')))
|
|
articles = ans.setdefault(section, [])
|
|
articles.append({'title': title, 'url': url})
|
|
for gird in soup.findAll(attrs={'class':'GridPromoTile'}):
|
|
for article in soup.findAll('article'):
|
|
a = article.find('a')
|
|
url = a['href']
|
|
if url.startswith('/'):
|
|
url = 'https://www.nationalgeographic.com' + url
|
|
if '/graphics/' in url:
|
|
continue
|
|
section = self.tag_to_string(article.find(**classes('SectionLabel')))
|
|
title = self.tag_to_string(article.find(**classes('PromoTile__Title--truncated')))
|
|
articles = ans.setdefault(section, [])
|
|
articles.append({'title': title, 'url': url})
|
|
self.log(pformat(ans))
|
|
return list(ans.items())
|
|
|
|
def preprocess_raw_html(self, raw_html, url):
|
|
data = extract_json(raw_html)
|
|
return '\n'.join(article_parse(data))
|
|
|
|
def preprocess_html(self, soup):
|
|
for h2 in soup.findAll('h2'):
|
|
h2.name = 'h4'
|
|
for img in soup.findAll('img', src=True):
|
|
# for high res images use '?w=2000&h=2000'
|
|
img['src'] = img['src'] + '?w=600&h=600'
|
|
return soup
|
|
|
|
def populate_article_metadata(self, article, soup, first):
|
|
summ = soup.find(attrs={'class':'byline'})
|
|
if summ:
|
|
article.summary = self.tag_to_string(summ)
|
|
article.text_summary = self.tag_to_string(summ)
|