Recipe for Caijing Magazine (zh-CN)

[gzeric]

I've developed the code for Caijing, the bi-weekly magazine. It's arguably the best finance and economics magazine in China. People call it the Economist in China.
The recipe is still work in progress. For example, I cannot figure out how to get the cover url correctly.

import string, re
from calibre import strftime
from calibre.web.feeds.recipes import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import BeautifulSoup


class Caijing(BasicNewsRecipe):

    title       = 'Caijing Magazine'
    __author__  = 'Eric Chen'

    description = '''Bi-weekly Finance and Economics Review. Founded in 1998, the fortnightly CAIJING 
                 Magazine has firmly established itself as a news authority and leading voice for 
                 business and financial issues in China. 
                 CAIJING Magazine closely tracks the most important aspects of China's economic reforms, 
                 developments and policy changes, as well as major events in the capital markets. It also 
                 offers a broad international perspective through first-hand reporting on international 
                 political and economic issues.
                 CAIJING Magazine is China's most widely read business and finance magazine, with a 
                 circulation of 225,000 per issue. It boasts top-level readers from government, business
                 and academic circles. '''
    language = 'zh'
    category = 'news, China'
    encoding = 'UTF-8'
    timefmt = ' [%a, %d %b, %Y]'
    needs_subscription = True

    remove_tags = [dict(attrs={'class':['topad', 'nav', 'searchbox', 'connav',
		'mbx', 'bianji', 'bianji bj', 'lnewlist', 'rdtj', 'loadComment', 
		'conr', 'bottom', 'bottomcopyr', 'emaildy', 'rcom', 'allcontent']}),
                dict(name=['script', 'noscript', 'style'])]
    no_stylesheets = True
    remove_javascript = True
    current_issue_url = ""
    current_issue_cover = ""


    def get_browser(self):
        br = BasicNewsRecipe.get_browser()
        if self.username is not None and self.password is not None:
            br.open('http://service.caijing.com.cn/usermanage/login')
            br.select_form(name='mainLoginForm')
            br['username'] = self.username
            br['password'] = self.password
            br.submit()
        return br

    def parse_index(self):
        global current_issue_cover
	articles = []
        soup0 = self.index_to_soup('http://magazine.caijing.com.cn/2011/cjindex2011/')
	div = soup0.find('div', attrs={'class':'fmcon'})
        link = div.find('a', href=True)
        current_issue_url = link['href']

    	soup = self.index_to_soup(current_issue_url)

	for div_cover in soup.findAll('img', {'src' : re.compile('.')}):
		if re.search('\d{4}-\d{2}-\d{2}', div_cover['src']):
			current_issue_cover = div_cover['src']

	feeds = []
        for section in soup.findAll('div', attrs={'class':'cebd'}):
	     section_title = self.tag_to_string(section.find('div', attrs={'class':'ceti'}))	     
	     articles = []
	     for post in section.findAll('a', href=True):
		 if re.search('\d{4}-\d{2}-\d{2}', post['href']):
                 	date = re.search('\d{4}-\d{2}-\d{2}', post['href']).group(0)
		 	id = re.search('\d{9}', post['href']).group(0)
		 	url = re.sub(r'\d.*', 'templates/inc/chargecontent2.jsp?id=', post['href'])
                 	url = url + id + '&time=' + date + '&cl=106&page=all'

		 	title = self.tag_to_string(post)
                    	articles.append({'title':title, 'url':url, 'date':date})

	     if articles:
                 feeds.append((section_title, articles))
        return feeds

    def get_cover_url(self):
	global current_issue_cover
	cover = current_issue_cover
        return cover

[forceps]

It maybe related to Beautifulsoup, the section contain cover is:

<div class=bigimg><em><img height=590 alt=财经杂志 src="http://img.caijing.com.cn/2011-03-14/110664866.jpg" width=459 border=0></em></div>

feed it to Beautifulsoup, the result is:

<div class="bigimg"><em><img height="590" alt="" /></em></div>

therefore u may need bypass soup and grab the cover url from current frontpage directly.

I would suggest add something like:
from contextlib import closing
br = BasicNewsRecipe.get_browser()
with closing(br.open(current_issue_url)) as f:

frontpage_html = f.read()

cover_urls = re.findall(r'<div class=bigimg(.*?)</div>', frontpage_html)
div_cover = re.findall(r'src="(.*?)"', str(cover_urls))[0]

[unpeople]

I tried your Caijing recipe at Calibre but was not successful. What username and password is required? I do have a registered user account with Caijing.