calibre/recipes/mit_technology_review.recipe

#!/usr/bin/env python
from __future__ import unicode_literals

__license__ = 'GPL v3'
__copyright__ = '2015 Michael Marotta <mikefm at gmail.net>'
# Written April 2015
# Last edited 08/2022
'''
technologyreview.com
'''
from calibre.web.feeds.news import BasicNewsRecipe, prefixed_classes
from collections import OrderedDict
import re

def absurl(x):
    if x.startswith('//'):
        x = 'http:' + x
    elif not x.startswith('http'):
        x = "http://www.technologyreview.com" + x
    return x


def classes(classes):
    q = frozenset(classes.split(' '))
    return dict(
        attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)}
    )


class MitTechnologyReview(BasicNewsRecipe):

    title = 'MIT Technology Review Magazine'
    __author__ = 'Michael Marotta, revised by unkn0wn'
    description = (
        'Bi-monthly magazine version of MIT Technology Review.'
        ' This is different than the recipe named simply "Technology Review"'
        ' which downloads the rss feed with daily articles from the website.'
    )
    INDEX = 'http://www.technologyreview.com/magazine/'
    language = 'en'
    encoding = 'utf-8'
    tags = 'news, technology, science'
    no_stylesheets = True
    remove_empty_feeds = True
    remove_attributes = ['height', 'width', 'style', 'padding', 'padding-top']
    delay = 1
    masthead_url = 'https://wp-preprod.technologyreview.com/wp-content/uploads/2021/08/Screen-Shot-2021-08-20-at-11.11.12-AM-e1629473232355.png'
    extra_css = '''
        #pub-d{font-size:small;}
        #cre-d{font-size:xx-small; text-align:center; color:gray;}
        #cap-d{font-size:small; text-align:center;}
        blockquote{text-align:center; color:#404040;}
        em { color:#202020;}
    '''
    keep_only_tags = [
        prefixed_classes('contentHeader contentArticleHeader contentBody')
    ]
    remove_tags = [
        dict(name="aside"),
        dict(name="svg"),
        prefixed_classes(
            'image__placeholder sliderAd__wrapper eyebrow__wrap-- screen-reader-text'
        ),
    ]

    def get_cover_url(self):
        soup = self.index_to_soup('https://www.technologyreview.com/')
        if script := soup.find('script', id='preload'):
            link = re.search('(https\S+?front_cover\S+?(jpg|png))', self.tag_to_string(script))
            return link.group(1) + '?fit=572,786'

    def parse_index(self):
        soup = self.index_to_soup(self.INDEX)
        issue = soup.find(attrs={'class':lambda x: x and x.startswith('magazineHero__title')})
        time = soup.find(attrs={'class': lambda x: x and x.startswith('magazineHero__date')})
        self.title = 'MIT Tech Review ' + self.tag_to_string(issue)
        self.timefmt = ' [' + self.tag_to_string(time) + ']'
        self.log('Downloading issue: ', self.timefmt)

        # parse articles
        feeds = OrderedDict()

        classNamePrefixes = [
            "magazineHero__letter--", "teaserItem__title", "teaserItem--aside__title"
        ]
        for div in soup.findAll(
            attrs={
                'class':
                lambda x: any(x.startswith(prefix) for prefix in classNamePrefixes)
                if x else False
            }
        ):
            articles = []
            a = div.find('a', href=True)
            title = self.tag_to_string(a).strip()
            href = absurl(a['href'])
            desc = ''
            section_title = 'Letter from the editor'
            d = div.findParent(
                attrs={
                    'class':
                    lambda z: z and z.
                    startswith(('teaserItem__wrapper', 'teaserItem--aside__wrapper'))
                }
            )
            if d:

                excerpt = d.find(
                        attrs={
                            'class':
                            lambda x: x and x.startswith(
                                ('teaserItem__excerpt', 'teaserItem--aside__excerpt')
                            )
                        }
                    )
                if excerpt:
                    desc = self.tag_to_string(excerpt).strip()

                sec = d.find(
                    attrs={
                        'class': lambda x: x and x.startswith('teaserItem__eyebrowText')
                    }
                )
                if sec:
                    section_title = self.tag_to_string(sec).replace('Categorized in ',
                                                            '').strip()

            if not href or not title:
                continue

            self.log(section_title)
            self.log('\t', title)
            self.log('\t', desc)
            self.log('\t\t', href)

            articles.append({'title': title, 'url': href, 'description': desc})
            if articles:
                if section_title not in feeds:
                    feeds[section_title] = []
                feeds[section_title] += articles
        ans = [(key, val) for key, val in feeds.items()]
        return ans

    def preprocess_html(self, soup):
        for bq in soup.findAll('blockquote'):
            for strong in bq.findAll('strong'):
                strong.name = 'div'
        for date in soup.findAll(
            attrs={
                'class':
                lambda x: x and x.
                startswith(('contentArticleHeader__publishDate', 'byline__wrapper'))
            }
        ):
            date['id'] = 'pub-d'
            for li in date.findAll(('li', 'ul')):
                li.name = 'span'
        for cap in soup.findAll('figcaption'):
            cap['id'] = 'cap-d'
        for credit in soup.findAll(
            attrs={
                'class':
                lambda x: x and x.startswith(('image__credit', 'image-credit'))
            }
        ):
            credit['id'] = 'cre-d'
        for img in soup.findAll(srcset=True):
            img['src'] = absurl(img['srcset'].split()[0])
            del img['srcset']
        for img in soup.findAll('img', attrs={'src': True}):
            img['src'] = img['src'].split('?')[0] + '?w=800'
        return soup