calibre/recipes/wsj.recipe

from calibre.web.feeds.jsnews import JavascriptRecipe
from calibre.web.jsbrowser.browser import NotAFile

def CSSSelect(expr):
    expr = {
        'div.whatsNews-simple': '''descendant-or-self::div[@class and contains(concat(' ', normalize-space(@class), ' '), ' whatsNews-simple ')]''',
        'a.mjLinkItem[href]': '''descendant-or-self::a[@class and contains(concat(' ', normalize-space(@class), ' '), ' mjLinkItem ') and (@href)]''',
        '.meta_sectionName': '''descendant-or-self::*[@class and contains(concat(' ', normalize-space(@class), ' '), ' meta_sectionName ')]''',
        'p':'descendant-or-self::p',
        'div.whatsNews-simple.whatsNews-itp': '''descendant-or-self::div[@class and contains(concat(' ', normalize-space(@class), ' '), ' whatsNews-simple ') and (@class and contains(concat(' ', normalize-space(@class), ' '), ' whatsNews-itp '))]''',
        'a[href]': 'descendant-or-self::a[@href]',
        'span.date-date':"descendant-or-self::span[@class and contains(concat(' ', normalize-space(@class), ' '), ' date-date ')]",
        'div.itpSectionHeaderPdf a[href]': "descendant-or-self::div[@class and contains(concat(' ', normalize-space(@class), ' '), ' itpSectionHeaderPdf ')]/descendant-or-self::*/a[@href]",
        'div.itpHeader ul.tab a[href]': "descendant-or-self::div[@class and contains(concat(' ', normalize-space(@class), ' '), ' itpHeader ')]/descendant-or-self::*/ul[@class and contains(concat(' ', normalize-space(@class), ' '), ' tab ')]/descendant-or-self::*/a[@href]",

    }[expr]
    from lxml.etree import XPath
    return XPath(expr)


class WSJ(JavascriptRecipe):

    title = 'The Wall Street Journal'
    __author__ = 'Kovid Goyal'
    description = 'News and current affairs'
    language = 'en'

    compress_news_images = True
    compress_news_images_auto_size = 7
    max_articles_per_feed = 1000
    timefmt = ' [%a, %b %d, %Y]'
    no_stylesheets = True
    ignore_duplicate_articles = {'url'}
    remove_attributes = ['style', 'data-scrim']
    needs_subscription = True

    keep_only_tags = (
        'h1',  # 'h2.subhead', 'h2.subHed.deck',
        'span[itemprop=author][rel=author]',
        'article#article-contents', 'article#articleBody',
        'div#article_story_body',
        # Parallax formatting
        'div#ncTitleArea', 'section.nc-exp-artbody',
        # Error conditions, login required and page not found
        'div#snippet-ad-login', 'div.errorNotFound',
    )

    remove_tags = (
        '.insetButton', '.insettipBox', '.author-info', '.media-object-video',
        '.article_tools', 'span[data-country-code][data-ticker-code]',
        'div.nc-exp-artmeta',
    )

    def do_login(self, br, username, password):
        br.visit(
            'https://id.wsj.com/access/pages/wsj/us/login_standalone.html?mg=com-wsj', timeout=120)  # noqa
        f = br.select_form(nr=0)
        f['username'] = username
        f['password'] = password
        br.submit(timeout=120)

    def preprocess_stage2(self, article, browser, url, recursion_level):
        # Slideshow and expandable images need to be processed here to
        # set the src attribute correctly
        found = 0
        for img in browser.css_select('img[data-in-base-data-lazy]', all=True):
            img.setAttribute('src', img.attribute('data-in-base-data-lazy'))
            found += 1
        for img in browser.css_select('img[data-enlarge]', all=True):
            img.setAttribute('src', img.attribute('data-enlarge'))
            found += 1
        if found:
            self.log.debug('Found %d dynamic images in:' % found, url)

    def get_publication_data(self, browser):
        return self.get_wsj_index(browser)

    def abs_wsj_url(self, href):
        if not href.startswith('http'):
            href = 'http://online.wsj.com' + href
        return href

    def wsj_find_articles(self, url):
        root = self.index_to_soup(url)

        for x in CSSSelect('div.whatsNews-simple')(root):
            x.getparent().remove(x)

        articles = []

        for a in CSSSelect('a.mjLinkItem[href]')(root):
            container = a.xpath('ancestor::li')
            meta = CSSSelect('.meta_sectionName')(a)
            if meta:
                meta = meta[0]
                meta.getparent().remove(meta)
                meta = self.tag_to_string(meta)
            title = self.tag_to_string(a)
            if meta:
                title += ' [%s]' % meta
            url = self.abs_wsj_url(a.get('href'))
            desc = ''
            if container:
                for p in CSSSelect('p')(container[0]):
                    desc = self.tag_to_string(p)
                    if 'Subscriber Content' not in desc:
                        break

            articles.append({'title': title, 'url': url,
                             'description': desc, 'date': ''})

            self.log('\tFound article:', title)
            self.log('\t\t', desc)
        return articles

    def wsj_find_wn_articles(self, url):
        root = self.index_to_soup(url)
        articles = []

        whats_news = CSSSelect('div.whatsNews-simple.whatsNews-itp')(root)
        if whats_news:
            for a in CSSSelect('a[href]')(whats_news[-1]):
                if '/articles/' not in a.get('href', ''):
                    continue
                container = a.xpath('ancestor::p')
                for meta in CSSSelect('.meta_sectionName')(a):
                    meta.getparent().remove(meta)
                title = self.tag_to_string(a).strip()
                url = self.abs_wsj_url(a.get('href'))
                desc = ''
                if container:
                    desc = self.tag_to_string(container[0])

                articles.append({'title': title, 'url': url,
                                 'description': desc, 'date': ''})

                self.log('\tFound WN article:', title)
                self.log('\t\t', desc)

        return articles

    def wsj_add_feed(self, feeds, title, url):
        self.log('Found section:', title)
        try:
            if url.endswith('whatsnews'):
                articles = self.wsj_find_wn_articles(url)
            else:
                articles = self.wsj_find_articles(url)
        except:
            articles = []
        if articles:
            feeds.append((title, articles))

    def get_wsj_index(self, browser):
        # return self.test_wsj_index()
        ans = {}
        root = self.index_to_soup('http://online.wsj.com/itp')
        for span in CSSSelect('span.date-date')(root):
            if span.text:
                self.timefmt = span.text
                break
        for a in CSSSelect('div.itpSectionHeaderPdf a[href]')(root):
            href = a.get('href')
            try:
                ans['cover'] = browser.download_file(href)
            except NotAFile:
                break
            break

        feeds = ans['index'] = []
        for a in CSSSelect('div.itpHeader ul.tab a[href]')(root):
            if '/itp/' not in a.get('href', ''):
                continue
            pageone = a.get('href').endswith('pageone')
            if pageone:
                title = 'Front Section'
                url = self.abs_wsj_url(a.get('href'))
                self.wsj_add_feed(feeds, title, url)
                title = "What's News"
                url = url.replace('pageone', 'whatsnews')
                self.wsj_add_feed(feeds, title, url)
            else:
                title = self.tag_to_string(a)
                url = self.abs_wsj_url(a.get('href'))
                self.wsj_add_feed(feeds, title, url)
        return ans

    def test_wsj_index(self):
        return {'index': [
            ('Testing', [
                {'title': 'Article One',
                 'url': 'http://online.wsj.com/articles/the-end-of-the-impulse-shopper-1416872108'},  # noqa
                {'title': 'Article Two',
                 'url': 'http://online.wsj.com/articles/ferguson-police-officer-not-charged-in-black-teens-shooting-1416882438'},  # noqa
                {'title': 'Article Three',
                 'url': 'http://online.wsj.com/article/SB10634695869867284248804580297251334393676.html'},  # noqa
            ]),
        ]}