Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-07 18:24:30 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix remaining issues in Newsweek recipe

Browse Source
This commit is contained in:
Kovid Goyal 2014-06-11 08:20:53 +05:30
parent b7e4c64da9
commit 3038bb58d4
1 changed files with 12 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
recipes/newsweek.recipe
View File

@ -7,8 +7,8 @@ def CSSSelect(expr):
return XPath(HTMLTranslator().css_to_xpath(expr)) return XPath(HTMLTranslator().css_to_xpath(expr))
BASE = 'http://www.newsweek.com' BASE = 'http://www.newsweek.com'
def href_to_url(a): def href_to_url(a, add_piano=False):
return BASE + a.get('href') + '?piano_t=1' return BASE + a.get('href') + ('?piano_t=1' if add_piano else '')
class Newsweek(JavascriptRecipe): class Newsweek(JavascriptRecipe):
@ -25,6 +25,7 @@ class Newsweek(JavascriptRecipe):
'meta', '.block-openadstream', '.block-ibtmedia-social', '.issue-next', 'meta', '.block-openadstream', '.block-ibtmedia-social', '.issue-next',
'.most-popular', '.ibt-media-stories', '.user-btn-group', '.most-popular', '.ibt-media-stories', '.user-btn-group',
'#taboola-below-main-column', '.trc_related_container', '#taboola-below-main-column', '.trc_related_container',
'#block-nw-magazine-magazine-more-from-issue', '.block-ibtmedia-top-stories',
] ]
LOGIN = 'https://bar.piano-media.com/lite/authent/login//custom/newsweek/?service_id=25&loc=http%3A%2F%2Fwww.newsweek.com%2F' # noqa LOGIN = 'https://bar.piano-media.com/lite/authent/login//custom/newsweek/?service_id=25&loc=http%3A%2F%2Fwww.newsweek.com%2F' # noqa
@ -41,7 +42,7 @@ class Newsweek(JavascriptRecipe):
root = self.index_to_soup(browser.html) root = self.index_to_soup(browser.html)
for a in CSSSelect('nav.main-menu a[href]')(root): for a in CSSSelect('nav.main-menu a[href]')(root):
if a.text and a.text.strip() == 'This Week\'s Edition': if a.text and a.text.strip() == 'This Week\'s Edition':
return self.get_newsweek_publication_data(browser, href_to_url(a)) return self.get_newsweek_publication_data(browser, href_to_url(a, True))
def get_newsweek_publication_data(self, browser, url): def get_newsweek_publication_data(self, browser, url):
root = self.index_to_soup(url) root = self.index_to_soup(url)
@ -97,17 +98,24 @@ class Newsweek(JavascriptRecipe):
ans['index'] = sections ans['index'] = sections
return ans return ans
def load_complete(self, browser, url, recursion_level):
browser.wait_for_element('div.article-body')
return browser.load_completed # This is needed to allow the parallax images to load
def preprocess_stage1(self, article, browser, url, recursion_level): def preprocess_stage1(self, article, browser, url, recursion_level):
# Parallax images in the articles are loaded as background images # Parallax images in the articles are loaded as background images
# on <span> tags. Convert them to normal images. # on <span> tags. Convert them to normal images.
for span in browser.css_select('span.parallax-image', all=True): for span in browser.css_select('span.parallax-image', all=True):
bg = unicode(span.styleProperty('background-image', span.InlineStyle)) bg = unicode(span.styleProperty('background-image', span.InlineStyle))
if bg: if bg:
url = bg.partition('(')[-1][:-1] url = bg.strip().partition('(')[-1][:-1]
span.appendInside('<img src="%s"></img>' % url) span.appendInside('<img src="%s"></img>' % url)
span.setAttribute('style', '') span.setAttribute('style', '')
browser.run_for_a_time(0.1) # This is needed to give the DOM time to update
def postprocess_html(self, article, root, url, recursion_level): def postprocess_html(self, article, root, url, recursion_level):
for x in root.xpath('//*[local-name()="body" and @style]'):
del x.attrib['style'] # body has a fixed height, which causes problems with epub viewers
for x in root.xpath('//*[@id="piano-root"]'): for x in root.xpath('//*[@id="piano-root"]'):
x.getparent().remove(x) x.getparent().remove(x)
return root return root