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

New recipe for Joop by kwetal

Browse Source
This commit is contained in:
Kovid Goyal 2010-01-16 21:15:23 -07:00
parent 3715fd26b2
commit 5bce3d10d3
6 changed files with 195 additions and 85 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
resources/images/news/joop.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 395 B

BIN
resources/images/news/nrcnext.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 1.7 KiB

BIN
resources/quick_start.epub Normal file
View File

Binary file not shown.

65
resources/recipes/fokkeensukke.recipe
View File

@ -1,23 +1,29 @@
#!/usr/bin/python
from calibre.web.feeds.news import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import Tag
from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag
class FokkeEnSukkeRecipe(BasicNewsRecipe) :
__license__ = 'GPL v3'
__author__ = 'kwetal'
language = 'nl'
description = u'Popular Dutch daily cartoon Fokke en Sukke'
country = 'NL'
version = 2
title = u'Fokke en Sukke'
no_stylesheets = True
# For reasons unknown to me the extra css is, on the cartoon pages, inserted in the <body> and not in the <head>. My reader (Sony PRS-600) has a serious issue
# with that: it treats it as content and displays it as is. Setting this property to empty solves this for me.
template_css = ''
INDEX = u'http://foksuk.nl'
publisher = u'Reid, Geleijnse & Van Tol'
category = u'News, Cartoons'
description = u'Popular Dutch daily cartoon Fokke en Sukke'
# This cover is not as nice as it could be, needs some work
#cover_url = 'http://foksuk.nl/content/wysiwyg/simpleimages/image350.gif'
conversion_options = {'comments': description, 'language': language, 'publisher': publisher}
no_stylesheets = True
extra_css = '''
body{font-family: verdana, arial, helvetica, geneva, sans-serif ; margin: 0em; padding: 0em;}
div.title {text-align: center; margin-bottom: 1em;}
'''
INDEX = u'http://foksuk.nl'
cover_url = 'http://foksuk.nl/content/wysiwyg/simpleimages/image350.gif'
keep_only_tags = [dict(name='div', attrs={'class' : 'cartoon'})]
@ -31,15 +37,14 @@ class FokkeEnSukkeRecipe(BasicNewsRecipe) :
links = index.findAll('a')
maxIndex = len(links) - 1
articles = []
for i in range(len(links)) :
# The first link does not interest us, as it points to no cartoon. A begin_at parameter in the range() function would be nice.
if i == 0 :
continue
# There can be more than one cartoon for a given day (currently either one or two). If there's only one, there is just a link with the dayname.
# If there are two, there are three links in sequence: <a>dayname</a> <a>1</a> <a>2</a>. In that case we're interested in the last two.
for i in range(1, len(links)) :
# There can be more than one cartoon for a given day (currently either one or two).
# If there's only one, there is just a link with the dayname.
# If there are two, there are three links in sequence: <a>dayname</a> <a>1</a> <a>2</a>.
# In that case we're interested in the last two.
if links[i].renderContents() in dayNames :
# If the link is not in daynames, we processed it already, but if it is, let's see if the next one has '1' as content
# If the link is not in daynames, we processed it already, but if it is, let's see
# if the next one has '1' as content
if (i + 1 <= maxIndex) and (links[i + 1].renderContents() == '1') :
# Got you! Add it to the list
article = {'title' : links[i].renderContents() + ' 1', 'date' : u'', 'url' : self.INDEX + links[i + 1]['href'], 'description' : ''}
@ -59,29 +64,31 @@ class FokkeEnSukkeRecipe(BasicNewsRecipe) :
return [[week, articles]]
def preprocess_html(self, soup) :
# This method is called for every page, be it cartoon or TOC. We need to process each in their own way
cartoon = soup.find('div', attrs={'class' : 'cartoon'})
if cartoon :
# It is a cartoon. Extract the title.
title = ''
img = soup.find('img', attrs = {'alt' : True})
if img :
title = img['alt']
# Using the 'extra_css' displays it in the <body> and not in the <head>. See comment at the top of this class. Setting the style this way solves that.
tag = Tag(soup, 'div', [('style', 'text-align: center; margin-bottom: 8px')])
tag = Tag(soup, 'div', [('class', 'title')])
tag.insert(0, title)
cartoon.insert(0, tag)
# I have not quite worked out why, but we have to throw out this part of the page. It contains the very same index we processed earlier,
# and Calibre does not like that too much. As far as I can tell it goes into recursion and the result is an empty eBook.
# We only want the cartoon, so throw out the index
select = cartoon.find('div', attrs={'class' : 'selectcartoon'})
if select :
select.extract()
return cartoon
else :
# It is a TOC. Just return the whole lot.
return soup
freshSoup = self.getFreshSoup(soup)
freshSoup.body.append(cartoon)
return freshSoup
def getFreshSoup(self, oldSoup):
freshSoup = BeautifulSoup('<html><head><title></title></head><body></body></html>')
if oldSoup.head.title:
freshSoup.head.title.append(self.tag_to_string(oldSoup.head.title))
return freshSoup

91
resources/recipes/joop.recipe Normal file
View File

@ -0,0 +1,91 @@
from calibre.web.feeds.news import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import Tag
import re
class JoopRecipe(BasicNewsRecipe):
__license__ = 'GPL v3'
__author__ = 'kwetal'
language = 'nl'
country = 'NL'
version = 1
title = u'Joop'
publisher = u'Vara'
category = u'News, Politics, Discussion'
description = u'Political blog from the Netherlands'
oldest_article = 7
max_articles_per_feed = 100
use_embedded_content = False
no_stylesheets = True
remove_javascript = True
keep_only_tags = []
keep_only_tags.append(dict(name = 'div', attrs = {'class': 'author_head clearfix photo'}))
keep_only_tags.append(dict(name = 'h2', attrs = {'class': 'columnhead smallline'}))
keep_only_tags.append(dict(name = 'div', attrs = {'class': re.compile('article.*')}))
extra_css = '''
body {font-family: verdana, arial, helvetica, geneva, sans-serif;}
img {margin-right: 0.4em;}
h3 {font-size: medium; font-style: italic; font-weight: normal;}
h2 {font-size: xx-large; font-weight: bold}
sub {color: #666666; font-size: x-small; font-weight: normal;}
div.joop_byline {font-size: large}
div.joop_byline_job {font-size: small; color: #696969;}
div.joop_date {font-size: x-small; font-style: italic; margin-top: 0.6em}
'''
INDEX = 'http://www.joop.nl'
conversion_options = {'comments': description, 'tags': category, 'language': language,
'publisher': publisher}
def parse_index(self):
sections = ['Politiek', 'Wereld', 'Economie', 'Groen', 'Media', 'Leven', 'Show', 'Opinies']
soup = self.index_to_soup(self.INDEX)
answer = []
div = soup.find('div', attrs = {'id': 'footer'})
for section in sections:
articles = []
h2 = div.find(lambda tag: tag.name == 'h2' and tag.renderContents() == section)
if h2:
ul = h2.findNextSibling('ul', 'linklist')
if ul:
for li in ul.findAll('li'):
title = self.tag_to_string(li.a)
url = self.INDEX + li.a['href']
articles.append({'title': title, 'date': None, 'url': url, 'description': ''})
answer.append((section, articles))
return answer
def preprocess_html(self, soup):
div = soup.find('div', 'author_head clearfix photo')
if div:
h2 = soup.find('h2')
if h2:
h2.name = 'div'
h2['class'] = 'joop_byline'
span = h2.find('span')
if span:
span.name = 'div'
span['class'] = 'joop_byline_job'
div.replaceWith(h2)
h2 = soup.find('h2', attrs = {'class': 'columnhead smallline'})
if h2:
txt = None
span = h2.find('span', 'info')
if span:
txt = span.find(text = True)
div = Tag(soup, 'div', attrs = [('class', 'joop_date')])
div.append(txt)
h2.replaceWith(div)
return soup

106
resources/recipes/ncrnext.recipe
View File

@ -1,29 +1,38 @@
from calibre.web.feeds.news import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import BeautifulSoup
from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag
class NrcNextRecipe(BasicNewsRecipe):
__license__ = 'GPL v3'
__author__ = 'kwetal'
version = 1
language = 'nl'
description = u'Dutch newsblog from the Dutch daily newspaper nrcnext.'
country = 'NL'
version = 2
title = u'nrcnext'
publisher = u'NRC Media'
category = u'News, Opinion, the Netherlands'
description = u'Dutch newsblog from the Dutch daily newspaper nrcnext.'
conversion_options = {'comments': description, 'language': language, 'publisher': publisher}
no_stylesheets = True
template_css = ''
remove_javascript = True
# I want to do some special processing on the articles. I could not solve it with the 'extra_css' property . So we do it the hard way.
keep_only_tags = [dict(name='div', attrs={'id' : 'main'})]
# If that's overkill for you comment out the previous line and uncomment the next. Then get rid of the preprocess_html() method.
#keep_only_tags = [dict(name='div', attrs={'class' : 'post'}), dict(name='div', attrs={'class' : 'vlag'}) ]
remove_tags = [dict(name = 'div', attrs = {'class' : 'meta'}),
dict(name = 'div', attrs = {'class' : 'datumlabel'}),
dict(name = 'ul', attrs = {'class' : 'cats single'}),
dict(name = 'ul', attrs = {'class' : 'cats onderwerpen'}),
dict(name = 'ul', attrs = {'class' : 'cats rubrieken'})]
remove_tags = []
remove_tags.append(dict(name = 'div', attrs = {'class' : 'meta'}))
remove_tags.append(dict(name = 'div', attrs = {'class' : 'datumlabel'}))
remove_tags.append(dict(name = 'ul', attrs = {'class' : 'cats single'}))
remove_tags.append(dict(name = 'ul', attrs = {'class' : 'cats onderwerpen'}))
remove_tags.append(dict(name = 'ul', attrs = {'class' : 'cats rubrieken'}))
use_embedded_content = False
extra_css = '''
body {font-family: verdana, arial, helvetica, geneva, sans-serif; text-align: left;}
p.wp-caption-text {font-size: x-small; color: #666666;}
h2.sub_title {font-size: medium; color: #696969;}
h2.vlag {font-size: small; font-weight: bold;}
'''
def parse_index(self) :
# Use the wesbite as an index. Their RSS feeds can be out of date.
@ -44,10 +53,11 @@ class NrcNextRecipe(BasicNewsRecipe):
# Find the links to the actual articles and rember the location they're pointing to and the title
a = post.find('a', attrs={'rel' : 'bookmark'})
href = a['href']
title = a.renderContents()
title = self.tag_to_string(a)
if index == 'columnisten' :
# In this feed/page articles can be written by more than one author. It is nice to see their names in the titles.
# In this feed/page articles can be written by more than one author.
# It is nice to see their names in the titles.
flag = post.find('h2', attrs = {'class' : 'vlag'})
author = flag.contents[0].renderContents()
completeTitle = u''.join([author, u': ', title])
@ -71,44 +81,46 @@ class NrcNextRecipe(BasicNewsRecipe):
return answer
def preprocess_html(self, soup) :
# This method is called for every page, be it cartoon or TOC. We need to process each in their own way
if soup.find('div', attrs = {'id' : 'main', 'class' : 'single'}) :
# It's an article, find the interesting part
if soup.find('div', attrs = {'id' : 'main', 'class' : 'single'}):
tag = soup.find('div', attrs = {'class' : 'post'})
if tag :
# And replace any links with their text, so they don't show up underlined on my reader.
for link in tag.findAll('a') :
link.replaceWith(link.renderContents())
if tag:
h2 = tag.find('h2', 'vlag')
if h2:
new_h2 = Tag(soup, 'h2', attrs = [('class', 'vlag')])
new_h2.append(self.tag_to_string(h2))
h2.replaceWith(new_h2)
else:
h2 = tag.find('h2')
if h2:
new_h2 = Tag(soup, 'h2', attrs = [('class', 'sub_title')])
new_h2.append(self.tag_to_string(h2))
h2.replaceWith(new_h2)
# Slows down my Sony reader; feel free to comment out
for movie in tag.findAll('span', attrs = {'class' : 'vvqbox vvqvimeo'}) :
h1 = tag.find('h1')
if h1:
new_h1 = Tag(soup, 'h1')
new_h1.append(self.tag_to_string(h1))
h1.replaceWith(new_h1)
# Slows down my reader.
for movie in tag.findAll('span', attrs = {'class' : 'vvqbox vvqvimeo'}):
movie.extract()
for movie in tag.findAll('span', attrs = {'class' : 'vvqbox vvqyoutube'}) :
for movie in tag.findAll('span', attrs = {'class' : 'vvqbox vvqyoutube'}):
movie.extract()
homeMadeSoup = BeautifulSoup('<html><head></head><body></body></html>')
body = homeMadeSoup.find('body')
body.append(tag)
return homeMadeSoup
else :
# This should never happen and other famous last words...
return soup
else :
# It's a TOC, return the whole lot.
return soup
def postproces_html(self, soup) :
# Should not happen, but it does. Slows down my Sony eReader
for img in soup.findAll('img') :
if img['src'].startswith('http://') :
img.extract()
# Happens for some movies which we are not able to view anyway
for iframe in soup.findAll('iframe') :
if iframe['src'].startswith('http://') :
for iframe in tag.findAll('iframe') :
iframe.extract()
fresh_soup = self.getFreshSoup(soup)
fresh_soup.body.append(tag)
return fresh_soup
else:
# This should never happen and other famous last words...
return soup
def getFreshSoup(self, oldSoup):
freshSoup = BeautifulSoup('<html><head><title></title></head><body></body></html>')
if oldSoup.head.title:
freshSoup.head.title.append(self.tag_to_string(oldSoup.head.title))
return freshSoup